DEVELOPMENT AND IN VITRO EVALUATION OF A CLOBETASOL … · 2017-02-12 · 1. To develop, optimize...

DEVELOPMENT AND IN VITRO EVALUATION OF A CLOBETASOL 17-PROPIONATE

TOPICAL CREAM FORMULATION

A Thesis Submitted in

Fulfilment of the Requirements for the Degree of

MASTER OF SCIENCE (PHARMACY)

of

RHODES UNIVERSITY

by

Kasongo Wa Kasongo

January 2007

Faculty of Pharmacy

Rhodes University

Grahamstown

South Africa

ii

ABSTRACT

One of the primary contributing factors to the escalating costs of health care is the high cost of

innovator pharmaceutical products. As a consequence, health authorities in various countries and in

particular in the developing world have identified generic prescribing and generic substitution as

possible strategies to contain the escalating costs of health care provision. There is therefore a need

for formulation scientists in developing countries to invest more time in the research and

development of generic formulations.

Clobetasol 17-propionate (CP) generic cream formulations containing 0.05% w/w of the drug were

manufactured and characterized using in vitro testing. Formulation development studies were

preceded by the development and validation of an RP-HPLC with UV detection for the quantitation

and characterization of CP in innovator and generic cream formulations during formulation

development and assessment studies. Furthermore the in vitro release rates of CP release from

innovator and generic cream formulations were monitored using a validated in vitro release test

method developed in these studies.

The formulation of CP cream products was accomplished using a variety of commercially available

mixed primary emulsifiers, such as Estol® 1474, Ritapro® 200, Emulcire® 61 WL and Gelot® 64.

Successful formulations were selected based on their ability to remain physically stable

immediately after manufacture and for 24 hours after storage at room temperature (22˚C). Estol®

1474 was found to produce an unstable cream and was therefore not investigated further.

iii

The other three emulgents produced stable creams, but only the in vitro release profile of CP from a

cream manufactured to contain Gelot® 64 was found to be statistically similar to that of the

innovator formulation. Therefore the cream containing Gelot® 64 was selected as the most

appropriate prototype generic cream formulation and was characterized in vitro in terms of CP

content, viscosity, pH and in vitro release rate. Data generated from these studies were compared to

those of the innovator product, Dermovate® cream, using statistical methods.

The CP content, pH and in vitro release rate data of the CP formulation were similar to those of the

innovator product, however the intrinsic viscosity of Dermovate® cream was almost three (3) times

greater than the intrinsic viscosity of the test formulation developed using Gelot® 64.

The CP cream formulation developed in these studies was stored for 4 weeks at 40 ± 2˚C and 25 ±

5% RH in an incubator and the formulation was found to be stable. A formulation has been

developed and assessed and found to be suitable for use as a topical semi-solid dosage form for CP.

iv

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to the following people:

My supervisor, Prof R.B. Walker for giving me the opportunity to be part of his Research Group

and for his support, patience, guidance, understanding and assistance throughout the course of my

studies and during the preparation of this thesis and also for providing me with laboratory facilities

and financial assistance.

My co-supervisor, Dr M.F. Skinner for his availability and assistance in various ways.

The Dean and Head, Prof I. Kanfer and the staff of the Faculty of Pharmacy, for use of the facilities

in the Faculty and Prof I. Kanfer for providing me with financial assistance.

Mr T. Samkange and Mr. L. Purdon for their technical expertise, assistance and advice.

Gattefossé SAS (Saint-Priest Cedex, France) for their donation of excipients.

My father for his love, encouragement, motivation, financial assistance, understanding and for

always believing in me not only throughout the duration of this project, but also throughout my life.

My mother, for her unfailing love and support. My sisters, brothers and Papa Andre for their

understanding and encouragement.

My colleagues in the Biopharmaceutics Research Laboratory (BRG) for their congenial company,

support, and encouragement throughout my study period.

Adrienne C. Muller for her support, encouragement and understanding throughout the course of my

project and for showing interest in my research.

The Almighty God for giving me protection, strength and resolve to succeed throughout my life.

v

STUDY OBJECTIVES

Eczema or dermatitis is a common dermatological disorder affecting approximately one-third of a

given population and is the most common skin condition in the Republic of South Africa (RSA).

Topical corticosteroid formulations such as creams or ointments applied three (3) times daily are

considered the most appropriate therapy for the treatment of eczema. Topical corticosteroids are

however considered costly for the majority of patients and as a consequence the use of white

petrolatum and hydrogenated vegetable oil as supplemental therapy is usually required. The high

cost of therapy is a result of the use of innovator products of the super-potent topical corticosteroids

such as clobetasol 17-propionate (CP) that are required for the treatment of severe or chronic

eczema, especially of the hands and feet. The availability of generic formulations of innovator

products will make medicinal products more affordable and accessible to a wider population.

The objectives of this study were:

1. To develop, optimize and validate a simple, selective, sensitive, precise, accurate and linear

reversed phase high performance liquid chromatographic method that is suitable for the

quantitative analysis of CP in cream formulations and for the assessment of CP release from

topical formulations during in vitro testing.

2. To develop and validate a reliable, reproducible and discriminatory in vitro release test method

for use in formulation development studies to assess product quality and ensure batch-to-batch

consistency of topical formulations manufactured to contain 0.05% w/w CP.

3. To design and develop a generic version of Dermovate® cream and to evaluate the product in

terms of several in vitro performance characteristics.

4. To determine the stability of the CP cream formulation developed in these studies at elevated

temperatures.

vi

TABLE OF CONTENTS

ABSTRACT --------------------------------------------------------------------------------------------------------------- II

ACKNOWLEDGEMENTS ------------------------------------------------------------------------------------------- IV

STUDY OBJECTIVES-------------------------------------------------------------------------------------------------- V

LIST OF TABLES ---------------------------------------------------------------------------------------------------- XIII

LIST OF FIGURES ---------------------------------------------------------------------------------------------------- XV

CHAPTER ONE -----------------------------------------------------------------------------------------------------------1

CLOBETASOL 17-PROPIONATE -----------------------------------------------------------------------------------1

1.1. INTRODUCTION -----------------------------------------------------------------------------------------------------1

1.2. DESCRIPTION --------------------------------------------------------------------------------------------------------2

1.3. PHYSICOCHEMICAL PROPERTIES-----------------------------------------------------------------------------2 1.3.1. Solubility ------------------------------------------------------------------------------------------------------------------- 2

1.3.1.1. Overview-----------------------------------------------------------------------------------------------------2 1.3.1.2. Solubility studies--------------------------------------------------------------------------------------------3

1.3.1.2.1. Overview -----------------------------------------------------------------------------------------------3 1.3.1.2.2. Propylene glycol (PG) --------------------------------------------------------------------------------3 1.3.1.2.3. Acetonitrile (ACN)------------------------------------------------------------------------------------3

1.3.2. Dissociation constant (pKa) -------------------------------------------------------------------------------------------- 4 1.3.3. Partition coefficient ------------------------------------------------------------------------------------------------------ 4 1.3.4. Melting range-------------------------------------------------------------------------------------------------------------- 4 1.3.5. Optical rotation------------------------------------------------------------------------------------------------------------ 4 1.3.6. Stability --------------------------------------------------------------------------------------------------------------------- 5 1.3.7. Ultraviolet absorption spectrum--------------------------------------------------------------------------------------- 5 1.3.8. Infra-red Absorption Spectrum---------------------------------------------------------------------------------------- 6 1.3.9. Crystal Structure ---------------------------------------------------------------------------------------------------------- 7

1.4. STEREOCHEMISTRY AND STRUCTURE ACTIVITY RELATIONSHIP --------------------------------8

1.5. CLINICAL PHARMACOLOGY -----------------------------------------------------------------------------------9 1.5.1. Mode of action ------------------------------------------------------------------------------------------------------------ 9

1.5.1.1. Overview-----------------------------------------------------------------------------------------------------9 1.5.1.2. Anti-inflammatory action-------------------------------------------------------------------------------- 10 1.5.1.3. Immuno-suppressive effects----------------------------------------------------------------------------- 11 1.5.1.4. Anti-mitotic and vasoconstrictive effects-------------------------------------------------------------- 12

1.5.2. Indications-----------------------------------------------------------------------------------------------------------------12 1.5.3. Contra-indications -------------------------------------------------------------------------------------------------------12 1.5.4. Adverse Effects ----------------------------------------------------------------------------------------------------------13

vii

1.5.4.1. Overview--------------------------------------------------------------------------------------------------- 13 1.5.4.2. Local side effects ----------------------------------------------------------------------------------------- 13 1.5.4.3. Systemic side effects ------------------------------------------------------------------------------------- 14

1.5.5. High risk groups ---------------------------------------------------------------------------------------------------------14 1.5.5.1. Pregnancy-------------------------------------------------------------------------------------------------- 14 1.5.5.2. Lactation --------------------------------------------------------------------------------------------------- 15 1.5.5.3. Paediatric use---------------------------------------------------------------------------------------------- 15 1.5.5.4. Geriatric use ----------------------------------------------------------------------------------------------- 16

1.6. PHARMACOKINETICS------------------------------------------------------------------------------------------- 16 1.6.1. Dosage and Administration -------------------------------------------------------------------------------------------16 1.6.2. Absorption-----------------------------------------------------------------------------------------------------------------17 1.6.3. Distribution ---------------------------------------------------------------------------------------------------------------17 1.6.4. Elimination----------------------------------------------------------------------------------------------------------------18

1.7. CONCLUSIONS----------------------------------------------------------------------------------------------------- 18

CHAPTER TWO -------------------------------------------------------------------------------------------------------- 21

DEVELOPMENT, OPTIMIZATION AND VALIDATION OF AN HPLC METHOD FOR THE ANALYSIS OF CLOBETASOL 17-PROPIONATE IN SEMI-SOLID DOSAGE FORMS------------- 21

2.1. INTRODUCTION --------------------------------------------------------------------------------------------------- 21

2.2. PRINCIPLES OF RP-HPLC --------------------------------------------------------------------------------------- 22

2.3. METHOD DEVELOPMENT-------------------------------------------------------------------------------------- 24 2.3.1. Overview ------------------------------------------------------------------------------------------------------------------24 2.3.2. Experimental--------------------------------------------------------------------------------------------------------------26

2.3.2.1. Chemicals -------------------------------------------------------------------------------------------------- 26 2.3.2.2. Instrumentation ------------------------------------------------------------------------------------------- 26 2.3.2.3. UV detection of CP and internal standard------------------------------------------------------------- 27 2.3.2.4. Column selection ----------------------------------------------------------------------------------------- 28 2.3.2.5. Column efficiency ---------------------------------------------------------------------------------------- 31 2.3.2.6. Internal standard selection------------------------------------------------------------------------------- 34 2.3.2.7. Mobile phase selection----------------------------------------------------------------------------------- 35 2.3.2.8. Preparation of mobile phase----------------------------------------------------------------------------- 37 2.3.2.9. Preparation of stock solutions and calibration standards -------------------------------------------- 37 2.3.2.10. Effect of ACN concentration -------------------------------------------------------------------------- 38 2.3.2.11. Effect of flow rate --------------------------------------------------------------------------------------- 39 2.3.2.12. Optimal mobile phase composition and flow rate -------------------------------------------------- 40 2.3.2.13. Chromatographic conditions--------------------------------------------------------------------------- 42

2.4. METHOD VALIDATION ----------------------------------------------------------------------------------------- 42 2.4.1. Overview ------------------------------------------------------------------------------------------------------------------42 2.4.2. Linearity and Range-----------------------------------------------------------------------------------------------------43 2.4.3. Precision -------------------------------------------------------------------------------------------------------------------45

2.4.3.1. Repeatability ---------------------------------------------------------------------------------------------- 45 2.4.3.2. Intermediate precision------------------------------------------------------------------------------------ 46

viii

2.4.3.3. Reproducibility-------------------------------------------------------------------------------------------- 47 2.4.4. Accuracy-------------------------------------------------------------------------------------------------------------------47 2.4.5. Limit of quantitation (LOQ) and limit of detection (LOD)----------------------------------------------------49 2.4.6. Specificity and selectivity ---------------------------------------------------------------------------------------------50

2.4.6.1. Overview--------------------------------------------------------------------------------------------------- 50 2.4.6.2. Sample preparation --------------------------------------------------------------------------------------- 50 2.4.6.3. Extraction efficiency ------------------------------------------------------------------------------------- 53 2.4.6.4. Validation of the extraction procedure----------------------------------------------------------------- 53 2.4.6.5. Selectivity studies----------------------------------------------------------------------------------------- 54

2.4.7. Sample stability ----------------------------------------------------------------------------------------------------------56 2.4.7.1. Overview--------------------------------------------------------------------------------------------------- 56 2.4.7.2. Stability data analysis ------------------------------------------------------------------------------------ 57 2.4.7.3. Stability of stock solutions ------------------------------------------------------------------------------ 58 2.4.7.4. In-process sample stability ------------------------------------------------------------------------------ 61

2.5. METHOD REVALIDATION ------------------------------------------------------------------------------------- 62 2.5.1. Overview ------------------------------------------------------------------------------------------------------------------62 2.5.2. Linearity -------------------------------------------------------------------------------------------------------------------63 2.5.3. Precision -------------------------------------------------------------------------------------------------------------------64

2.5.3.1. Repeatability ---------------------------------------------------------------------------------------------- 64 2.5.3.2. Intermediate precision------------------------------------------------------------------------------------ 65 2.5.3.3. Reproducibility-------------------------------------------------------------------------------------------- 65

2.5.4. Accuracy-------------------------------------------------------------------------------------------------------------------66

2.6. APPLICATION OF THE ANALYTICAL METHOD --------------------------------------------------------- 66

2.7. CONCLUSIONS----------------------------------------------------------------------------------------------------- 67

CHAPTER THREE ----------------------------------------------------------------------------------------------------- 70

DEVELOPMENT AND VALIDATION OF AN IN VITRO TEST METHOD FOR THE ASSESSMENT OF CLOBETASOL 17-PROPIONATE RELEASE FROM TOPICAL CREAM FORMULATIONS ------------------------------------------------------------------------------------------------------ 70

3.1. INTRODUCTION --------------------------------------------------------------------------------------------------- 70

3.2. METHOD DEVELOPMENT-------------------------------------------------------------------------------------- 73 3.2.1. Overview ------------------------------------------------------------------------------------------------------------------73 3.2.2. Diffusion cell test system ----------------------------------------------------------------------------------------------73

3.2.2.1. Overview--------------------------------------------------------------------------------------------------- 73 3.2.2.2. Franz diffusion cell --------------------------------------------------------------------------------------- 75 3.2.2.3. Modified Franz diffusion cell --------------------------------------------------------------------------- 76 3.2.2.4. Selection of diffusion cell test system ----------------------------------------------------------------- 77

3.2.3. Number of samples------------------------------------------------------------------------------------------------------78 3.2.4. Sampling times -----------------------------------------------------------------------------------------------------------78 3.2.5. Temperature---------------------------------------------------------------------------------------------------------------81 3.2.6. Receptor medium --------------------------------------------------------------------------------------------------------82

3.2.6.1. Overview--------------------------------------------------------------------------------------------------- 82 3.2.6.2. Selection of receptor medium--------------------------------------------------------------------------- 83

ix

3.2.6.2.1. Aqueous systems ------------------------------------------------------------------------------------ 83 3.2.6.2.2. Water-immiscible systems ------------------------------------------------------------------------- 83 3.2.6.2.3. Water-miscible systems ---------------------------------------------------------------------------- 83

3.2.6.2.3.1. Overview---------------------------------------------------------------------------------------- 83 3.2.6.2.3.2. Alcohol/water mixtures ----------------------------------------------------------------------- 84 3.2.6.2.3.3. Propylene glycol/water mixtures ------------------------------------------------------------ 85

3.2.6.3. Saturation solubility -------------------------------------------------------------------------------------- 86 3.2.6.4. Preparation of the receptor medium-------------------------------------------------------------------- 89

3.2.7. Synthetic membranes ---------------------------------------------------------------------------------------------------89 3.2.7.1. Overview--------------------------------------------------------------------------------------------------- 89 3.2.7.2. Characteristics of membranes--------------------------------------------------------------------------- 90 3.2.7.3. Assessment of membranes ------------------------------------------------------------------------------ 91 3.2.7.4. Membrane selection -------------------------------------------------------------------------------------- 93

3.2.7.4.1. Silicone membrane---------------------------------------------------------------------------------- 94 3.2.7.4.2. Porous membranes ---------------------------------------------------------------------------------- 94

3.2.7.5. Membrane resistance ------------------------------------------------------------------------------------- 96 3.2.8. Sample Application -----------------------------------------------------------------------------------------------------98

3.2.8.1. Overview--------------------------------------------------------------------------------------------------- 98 3.2.8.2. Effects of sample application --------------------------------------------------------------------------- 98

3.2.9. Sample occlusion ------------------------------------------------------------------------------------------------------ 100 3.2.9.1. Overview--------------------------------------------------------------------------------------------------100 3.2.9.2. Effects of occlusion -------------------------------------------------------------------------------------101

3.2.10. Sample Analysis------------------------------------------------------------------------------------------------------ 102 3.2.11. Comparison of diffusion or release rate profiles-------------------------------------------------------------- 103 3.2.12. Optimal in vitro release test conditions------------------------------------------------------------------------- 104

3.3. METHOD VALIDATION ----------------------------------------------------------------------------------------105 3.3.1. Overview ---------------------------------------------------------------------------------------------------------------- 105 3.3.2. Changes in dosage strength ----------------------------------------------------------------------------------------- 106 3.3.3. Changes in composition---------------------------------------------------------------------------------------------- 107

3.3.2.3. Changes in viscosity-------------------------------------------------------------------------------------109 3.3.2.3.1. Overview --------------------------------------------------------------------------------------------109 3.3.2.3.2. Determination of viscosity ------------------------------------------------------------------------110 3.3.2.3.3. Effects of viscosity---------------------------------------------------------------------------------110

3.4. CONCLUSIONS----------------------------------------------------------------------------------------------------112

CHAPTER FOUR ------------------------------------------------------------------------------------------------------114

DEVELOPMENT AND IN VITRO CHARACTERIZATION OF CLOBETASOL 17-PROPIONATE CREAM FORMULATIONS-----------------------------------------------------------------------------------------114

4.1. INTRODUCTION --------------------------------------------------------------------------------------------------114

4.2. CREAM FORMULATIONS--------------------------------------------------------------------------------------117 4.2.1. Overview ---------------------------------------------------------------------------------------------------------------- 117 4.2.2. Instability mechanisms in creams---------------------------------------------------------------------------------- 120

4.2.2.1. Overview--------------------------------------------------------------------------------------------------120 4.2.2.2. Flocculation ----------------------------------------------------------------------------------------------121

x

4.2.2.3. Coalescence-----------------------------------------------------------------------------------------------122 4.2.2.4. Creaming or sedimentation-----------------------------------------------------------------------------122 4.2.2.5. Ostwald ripening-----------------------------------------------------------------------------------------123 4.2.2.6. Phase inversion-------------------------------------------------------------------------------------------124

4.2.3. Stabilization of creams ----------------------------------------------------------------------------------------------- 124 4.2.3.1. Surfactants ------------------------------------------------------------------------------------------------124 4.2.3.2. Mixed emulgents-----------------------------------------------------------------------------------------125 4.2.3.3. Theory of cream emulsification------------------------------------------------------------------------127 4.2.3.4. Hydrophilic-lipophilic balance (HLB)----------------------------------------------------------------130

4.3. EXPERIMENTAL--------------------------------------------------------------------------------------------------132 4.3.1. Characterization of CP creams ------------------------------------------------------------------------------------- 132

4.3.1.1. Overview--------------------------------------------------------------------------------------------------132 4.3.1.2. Assay of CP content -------------------------------------------------------------------------------------132 4.3.1.3. Intrinsic viscosity ----------------------------------------------------------------------------------------133 4.3.1.4. pH-determination ----------------------------------------------------------------------------------------133 4.3.1.5. In vitro release rate --------------------------------------------------------------------------------------134

4.3.2. Innovator product characterization -------------------------------------------------------------------------------- 135 4.3.2.1. Overview--------------------------------------------------------------------------------------------------135 4.3.2.2. Qualitative composition---------------------------------------------------------------------------------135 4.3.2.3. CP content ------------------------------------------------------------------------------------------------135 4.3.2.4. Intrinsic viscosity ----------------------------------------------------------------------------------------136 4.3.2.5. pH-determination ----------------------------------------------------------------------------------------136 4.3.2.6. In vitro release rate studies -----------------------------------------------------------------------------136

4.3.3. Generic product development -------------------------------------------------------------------------------------- 138 4.3.3.1. Overview--------------------------------------------------------------------------------------------------138 4.3.3.2. Excipients -------------------------------------------------------------------------------------------------139

4.3.3.2.1. Overview --------------------------------------------------------------------------------------------139 4.3.3.2.2. Clobetasol 17-propionate--------------------------------------------------------------------------139 4.3.3.2.3. Propylene glycol------------------------------------------------------------------------------------139 4.3.3.2.4. Sodium citrate --------------------------------------------------------------------------------------140 4.3.3.2.5. Citric acid -------------------------------------------------------------------------------------------140 4.3.3.2.6. Geleol®-----------------------------------------------------------------------------------------------140 4.3.3.2.7. Cetostearyl alcohol---------------------------------------------------------------------------------141 4.3.3.2.8. White beeswax--------------------------------------------------------------------------------------142 4.3.3.2.9. Chlorocresol ----------------------------------------------------------------------------------------142 4.3.3.2.10. Estol® 1474 ----------------------------------------------------------------------------------------143 4.3.3.2.11. Ritapro® 200---------------------------------------------------------------------------------------143 4.3.3.2.12. Emulcire® 61 WL ---------------------------------------------------------------------------------143 4.3.3.2.13. Gelot® 64-------------------------------------------------------------------------------------------143

4.3.4. Formulation composition -------------------------------------------------------------------------------------------- 144 4.3.5. Manufacturing methods ---------------------------------------------------------------------------------------------- 146

4.3.5.1. Overview--------------------------------------------------------------------------------------------------146 4.3.5.2. Aqueous phase -------------------------------------------------------------------------------------------146 4.3.5.3. Oil phase --------------------------------------------------------------------------------------------------146 4.3.5.4. Dispersed phase ------------------------------------------------------------------------------------------147 4.3.5.5. Drug phase------------------------------------------------------------------------------------------------147 4.3.5.6. Cream formulation---------------------------------------------------------------------------------------147

4.3.6. Preliminary studies ---------------------------------------------------------------------------------------------------- 149 4.5.7. In vitro release studies------------------------------------------------------------------------------------------------ 151

4.5.7.1. Overview--------------------------------------------------------------------------------------------------151

xi

4.5.7.2. Effects of Ritapro® 200 ---------------------------------------------------------------------------------151 4.5.7.3. Effects of Emulcire® 61 WL----------------------------------------------------------------------------153 4.5.7.4. Effects of Gelot® 64 -------------------------------------------------------------------------------------155

4.3.8. Generic product characterization ---------------------------------------------------------------------------------- 157 4.3.8.1. Overview--------------------------------------------------------------------------------------------------157 4.3.8.2. CP content ------------------------------------------------------------------------------------------------157 4.3.8.3. Intrinsic viscosity ----------------------------------------------------------------------------------------158 4.3.8.4. pH-determination ----------------------------------------------------------------------------------------161 4.3.8.5. In vitro release rate --------------------------------------------------------------------------------------161

4.4. CONCLUSIONS----------------------------------------------------------------------------------------------------164

CHAPTER FIVE -------------------------------------------------------------------------------------------------------167

STABILITY OF CLOBETASOL 17-PROPIONATE CREAMS --------------------------------------------167

5.1. INTRODUCTION --------------------------------------------------------------------------------------------------167

5.2. EXPERIMENTAL--------------------------------------------------------------------------------------------------171 5.2.1. Overview ---------------------------------------------------------------------------------------------------------------- 171 5.2.2. Stability study protocol----------------------------------------------------------------------------------------------- 172

5.2.2.1. Overview--------------------------------------------------------------------------------------------------172 5.2.2.2. Selection of batches -------------------------------------------------------------------------------------172 5.2.2.3. Number of batches---------------------------------------------------------------------------------------173 5.2.2.4. Container-closure system-------------------------------------------------------------------------------173 5.2.2.5. Sampling frequency -------------------------------------------------------------------------------------173 5.2.2.6. Sampling plan --------------------------------------------------------------------------------------------174 5.2.2.7. Test storage conditions----------------------------------------------------------------------------------174 5.2.2.8. Test specifications ---------------------------------------------------------------------------------------175 5.2.2.9. Product specifications -----------------------------------------------------------------------------------175 5.2.2.10. Methodology--------------------------------------------------------------------------------------------176

5.2.2.10.1. Test procedure-------------------------------------------------------------------------------------176 5.2.2.10.2. Organoleptic appeal ------------------------------------------------------------------------------177 5.2.2.10.3. CP content -----------------------------------------------------------------------------------------177 5.2.2.10.4. Intrinsic viscosity ---------------------------------------------------------------------------------177 5.2.2.10.5. pH-determination ---------------------------------------------------------------------------------177 5.2.2.10.6. In vitro release test--------------------------------------------------------------------------------177

5.2.2.11. Statistical evaluation -----------------------------------------------------------------------------------178 5.2.3. Results and discussion------------------------------------------------------------------------------------------------ 179

5.2.3.1. Qualitative analysis--------------------------------------------------------------------------------------179 5.2.3.2. Quantitative analysis ------------------------------------------------------------------------------------179 5.2.3.2.1. CP content ----------------------------------------------------------------------------------------------180 5.2.3.3. Intrinsic viscosity ----------------------------------------------------------------------------------------181 5.2.3.4. Apparent intrinsic pH -----------------------------------------------------------------------------------182 5.2.3.5. In vitro release rate testing -----------------------------------------------------------------------------183

5.3. CONCLUSIONS----------------------------------------------------------------------------------------------------186

xii

CHAPTER SIX ---------------------------------------------------------------------------------------------------------189

CONCLUSIONS--------------------------------------------------------------------------------------------------------189

APPENDIX ONE -------------------------------------------------------------------------------------------------------197

APPENDIX TWO ------------------------------------------------------------------------------------------------------198

APPENDIX THREE ---------------------------------------------------------------------------------------------------199

APPENDIX FOUR -----------------------------------------------------------------------------------------------------200

APPENDIX FIVE ------------------------------------------------------------------------------------------------------201

APPENDIX SIX---------------------------------------------------------------------------------------------------------208

APPENDIX SEVEN----------------------------------------------------------------------------------------------------215

REFERENCES----------------------------------------------------------------------------------------------------------217

xiii

LIST OF TABLES

Table 1.1. Solubility of CP............................................................................................................................... 3

Table 1.2. Major infra-red band assignment for CP......................................................................................... 7

Table 2.1. RP-HPLC methods used for the analysis of CP in semi-solid dosage forms ................................ 25

Table 2.2. Retention times of CP and potential internal standards................................................................. 34

Table 2.3. Chromatographic conditions for the analysis of CP...................................................................... 42

Table 2.4. Repeatability data for HPLC analysis of CP ................................................................................. 46

Table 2.5. Intermediate precision data for HPLC analysis of CP .................................................................. 47

Table 2.6. Accuracy data for HPLC analysis of CP (n = 3) ........................................................................... 48

Table 2.7. LOQ data for HPLC analysis of CP .............................................................................................. 50

Table 2.8. Extraction efficiency data following the extraction of CP from Dermovate® cream (n = 6) ........ 53

Table 2.9. Extraction efficiency following extraction of 600 µl of a CP solution (0.05% w/w) (n = 3)........ 54

Table 2.10. Repeatability data for the revalidation of the HPLC method for CP .......................................... 65

Table 2.11. Intermediate precision data for the revalidation of the HPLC method for CP............................ 65

Table 2.12. Accuracy data for the revalidation of the HPLC method for CP (n = 3)..................................... 66

Table 3.1. Solubility data for CP in various concentrations of PG at 32˚C (n = 3)........................................ 87

Table 3.2. Summary of the characteristics of synthetic membranes .............................................................. 91

Table 3.3. Cumulative percentage CP released after 72 hour (n = 3) ............................................................ 95

Table 3.4. In vitro release characteristics from four loading doses applied to the test membrane............... 100

Table 3.5. Summary of optimal in vitro release test conditions ................................................................... 105

Table 3.6. Effect of changes in the CP concentration on the total cumulative amount released and the

associated flux values (n = 3) ...................................................................................................... 106

Table 3.7. Effect of changes in formulation composition on flux and cumulative amount of CP released (n =

3).................................................................................................................................................. 108

Table 3.8. Cumulative amount of CP released and the average in vitro release rate (flux) from CP cream

formulations of different intrinsic viscosity (n = 3)..................................................................... 110

Table 4.1. Pharmacopoeial and commercially available emulsifying waxes [172, 173] ............................. 127

Table 4.2. Percentage composition of generic CP cream formulations developed and assessed in these

studies .......................................................................................................................................... 145

xiv

Table 4.3. Cumulative amount of CP released and the average in vitro release rate (flux) of CP from Batch

CP002 and Dermovate® cream (n = 6) ........................................................................................ 151

Table 4.4. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I) calculated using

Batch CP002 (test) and Dermovate® cream (reference) .............................................................. 152



Table 4.6. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I.) calculated for using

Batch CP003 (test) and Dermovate® cream (reference) .............................................................. 154



Table 4.8. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I) calculated using

Dermovate® cream (reference) and Batch CP004 (test) .............................................................. 156

Table 4.9. CP content of Dermovate® cream and Batches of generic cream formulations (n = 3) .............. 157

Table 4.10. Intrinsic viscosity readings for Dermovate® cream and generic cream products (n = 3) .......... 158

Table 4.11. pH readings for Dermovate® cream and generic cream formulations (n = 3) .......................... 161

Table 4.12. Cumulative amount of CP released and the average in vitro release rate (flux) of CP from

Dermovate® cream and generic cream formulations (n = 6) ..................................................... 162

Table 4.13. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I.) calculated using

Dermovate® cream (reference) and generic formulations (tests) .............................................. 163

Table 5.1. Specifications for qualitative parameters of the test cream formulation ..................................... 176

Table 5.2. Specifications for quantitative parameters of the test cream formulation ................................... 176

Table 5.3. Stability data generated for Batch CP004 after a four (4) week test period................................ 180

Table 5.4. The lower limit (L.L.) and upper limit (U.L.) of the confidence intervals (C.I.) calculated using

cream sample at week 0 (reference) and cream samples at weeks 1, 2, 3 and 4 (test) .............. 186

xv

LIST OF FIGURES

Figure 1.1. Chemical structure of clobetasol 17-propionate (C25H32ClFO5, MW = 467.0) ............................. 2

Figure 1.2. UV absorption spectrum of CP in ACN:water (50:50).................................................................. 6

Figure 1.3. IR absorption spectrum of CP (adapted from 22).......................................................................... 7

Figure 2.1. Typical chromatogram of a test mixture containing uracil (1), acetophenone (2), benzene (3),

toluene (4) and naphthalene (5) after separation on a 4 µm Nova-Pak® C18 (150 mm x 3.9 i.d.)

cartridge column.......................................................................................................................... 33

Figure 2.2. Typical chromatograms generated using a binary mixture of MeOH and water (A) and a binary

mixture of ACN and water (B) as mobile phase. ........................................................................ 36

Figure 2.3. Effect of ACN concentration on the retention times of BV and CP............................................ 38

Figure 2.4. Effect of mobile phase (50% v/v ACN) flow rate on the retention times of BV and CP ............ 40

Figure 2.5. Typical chromatogram of a mixture of the internal standard, betamethasone 17-valerate (BV)

and clobetasol 17-propionate (CP) using a mobile phase of 50% v/v ACN-water and a flow rate

of 1.0 ml/min ............................................................................................................................... 41

Figure 2.6. Calibration curve constructed for CP following least squares linear regression analysis of peak

height ratios of CP and IS versus concentration.......................................................................... 44

Figure 2.7. Schematic representation of the sample preparation procedure .................................................. 52

Figure 2.8. Typical chromatograms obtained following the analysis of a sample of placebo cream without

BV (A) and with BV (B) and Dermovate® cream without BV (C) and with BV (D) ................. 55

Figure 2.9. Interpretation of stability data, as described by Timm et al, [97]................................................ 58

Figure 2.10. Stability of CP in ACN at two different concentrations, stored at + 4ºC for 1, 2, 3, 7, and 14

days.............................................................................................................................................. 60

Figure 2.11. Stability of CP in 50% v/v propylene glycol/water stored at + 22ºC for 1, 2 and 3 days.......... 62

Figure 2.12. Calibration curve for CP following revalidation of the method ................................................ 64

Figure 2.13. In vitro release profile of CP ..................................................................................................... 67

Figure 3.1. Schematic representation of an original Franz diffusion cell apparatus (redrawn from 127)...... 76

Figure 3.2. Schematic representation of a modified Kenshary-Chien Franz glass diffusion cell (redrawn

from 127)..................................................................................................................................... 77

Figure 3.3. Schematic representation of a modified Franz cell multiple-cell drive unit ................................ 78

Figure 3.4. In vitro release of CP from Dermovate® cream (n = 6) ............................................................... 80

xvi

Figure 3.5. Least squares linear regression best fit line of the in vitro release profile of CP from Dermovate®

cream (n = 6) ............................................................................................................................... 81

Figure 3.6. In vitro release profile of CP from Dermovate® cream, using 30% v/v ethanol solution as

receptor medium (n = 6).............................................................................................................. 85

Figure 3.7. In vitro release profile of CP from Dermovate ® cream using 30% v/v PG solution as receptor

medium (n = 6) ............................................................................................................................ 86

Figure 3.8. Saturation solubility profile of CP in PG:water solutions of different proportions (n = 3) ......... 88

Figure 3.9. Effect of membrane type on the in vitro release of CP from Dermovate® cream (n = 3)............ 93

Figure 3.10. Comparison of in vitro release of CP from a 0.05% v/v CP solution and Dermovate® cream

using a 0.025 µm nitrocellulose membrane (n = 6)..................................................................... 97

Figure 3.11. Effect of sample loading on the in vitro release of CP from Dermovate® cream ...................... 99

Figure 3.12. Effect of occlusion on the in vitro release of CP from Dermovate® cream (n = 6).................. 101

Figure 3.13. CP calibration curve used to determine the amount of CP released in in vitro release

experiments ............................................................................................................................. 103

Figure 3.14. Effect of changes in CP content on the in vitro release rate of CP from an extemperaneous

cream base (n = 3) ................................................................................................................... 107

Figure 3.15. Effect of changes in formulation composition on CP in vitro release rate (n = 3). ................. 108

Figure 3.16. Effect of changes in the intrinsic viscosity of a formulation on the in vitro release rate of CP (n

= 3). ........................................................................................................................................... 111

Figure 4.1. Schematic representation of the principles of oil-in-water (o/w) and water-in-oil (w/o) emulsions

(redrawn from 181). .................................................................................................................. 119

Figure 4.2. Schematic representation of the mechanisms by which creams show instability (adapted from

184)............................................................................................................................................ 121

Figure 4.3. Illustration of the gel network theory (adapted from 171)......................................................... 128

Figure 4.4. In vitro release rate profile for CP from Dermovate® cream (n = 6) ......................................... 137

Figure 4.5. Schematic illustration of the manufacturing method for the CP cream formulation ................. 148

Figure 4.6. In vitro release rate profile for CP from Batch CP002 and Dermovate® cream ........................ 152



Figure 4.9. In vitro release rate profile for CP release from CP generic cream products and Dermovate®

cream ......................................................................................................................................... 162

xvii

Figure 5.1. Effects of stability test conditions on CP content of the cream formulation after storage at 40 ±

2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks............................................................................ 181

Figure 5.2. Effects of stability test conditions on the apparent intrinsic viscosity of the CP cream

formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks. ...................... 182

Figure 5.3. Effects of stability test conditions on the apparent intrinsic pH of the cream formulation after

storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks. .................................................. 183

Figure 5.4. Effects of stability test conditions on the in vitro release rates of CP from the cream formulation

after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks ........................................... 184

Figure 5.5. In vitro release rate profiles for CP from CP cream samples at times 0, 1, 2, 3 and 4 weeks after

storage ....................................................................................................................................... 185

1

CHAPTER ONE

CLOBETASOL 17-PROPIONATE

1.1. INTRODUCTION

It is generally believed that the introduction of hydrocortisone in the early 1950s, for the

treatment of patients with inflammatory conditions such as atopic dermatitis and other

eczematous eruptions was a milestone in topical therapy of dermatological conditions [1, 2].

Subsequently, various topical corticosteroids of increasing potency have been synthesised and

made available to dermatologists, whilst research has led toward the concomitant development of

vehicles that enhance the activity of the glucocorticoid class of drugs [3-5].

The potency levels of the corticosteroids can be measured by use of the human vasoconstrictor

[6] or skin blanching [7] assay, in which blanching is assessed in healthy volunteers after topical

application of a corticosteroid formulation [6-8]. The blanching assay has been a reliable

technique for the determination of the relative potency of topical corticosteroids, and has been

shown to be effective for the correlation of potency and clinical effectiveness [8]. Based on the

vasoconstrictor assay, topical corticosteroids are ranked according to their potency level, with the

weakest class, including hydrocortisone, labelled as class VII compounds and the most potent

steroids, including clobetasol 17-propionate, labelled as class I compounds [1, 2].

The chemical structure of clobetasol 17-propionate (CP) is depicted in Figure 1.1. CP is a class I

or super-potent synthetic di-halogenated analogue of prednisolone [1, 9-11]. CP is 1800 times

more potent than hydrocortisone when potency is measured using the human skin blanching

assay [1, 7] and it is currently the most potent topical corticosteroid available on the market [2, 3,

9, 10, 12-14]. Since 1973, CP has been used for the short-term treatment of patients with

inflammatory and pruritic manifestations of moderate-to-severe glucocorticoid-responsive

dermatoses [1, 10, 11]. CP is currently available as a 0.05% w/w CP formulation in a variety of

vehicles, including cream, ointment, gel, lotion and more recently foams [15-17].

2

1.2. DESCRIPTION

CP is 21-chloro-9α-fluoro-11β,17α-dihydroxy-16β-methylpregna-1,4-diene-3,20-dione 17-

propionate[18-21].

O

F

HOO

OO

Cl

A B

C D1

2

3

45

6

7

89

10

19 11

12

13

1415

16

17

18

20

21

Figure 1.1. Chemical structure of clobetasol 17-propionate (C25H32ClFO5, MW = 467.0)

CP occurs as a white, almost white or cream-coloured, crystalline powder [15, 18, 22] and is

odourless [10, 14]. CP contains not less than 97.0 percent and not more than 102.0 percent

C25H32ClFO5, calculated with reference to a dry reference standard substance [22, 23].

1.3. PHYSICOCHEMICAL PROPERTIES

1.3.1. Solubility

1.3.1.1. Overview

The solubility data of CP in water, ethanol and other organic solvents at room temperature

(22○C) are summarized in Table 1.1 [10, 14, 18, 20-22]. These data reveal that CP is practically

insoluble in water, is sparingly soluble in ether, but is soluble in ethanol and is freely soluble in

acetone, chloroform and dichloromethane [18, 20, 21, 23].

3

Table 1.1. Solubility of CP

Solvent Solubility Water 2 µg/ml Ethanol 10 mg/ml Ether 1 in 1000 Acetone 1 in 10 Chloroform 1 in 10 Dichloromethane 1 in 10

1.3.1.2. Solubility studies

1.3.1.2.1. Overview

The solubility of CP in propylene glycol (PG) and acetonitrile (ACN) has not yet been reported.

Consequently solubility studies of CP in PG and ACN were undertaken. Since PG and ACN

were used in the preparation of the receptor medium for use in in vitro release test studies

(Section 3.2.6.4., Chapter 3) and in the preparation of the mobile phase for use in RP-HPLC

studies (Section 2.3.2.8., Chapter 2) respectively, it was considered essential to evaluate the

solubility of CP in these two solvents.

1.3.1.2.2. Propylene glycol (PG)

The solubility of CP in PG was determined experimentally as described and reported in Section

3.2.6.3., and was found to be 8.55 ± 2.51 mg/ml (n = 3) (Table 3.1, Section 3.2.6.3, Chapter 3).

1.3.1.2.3. Acetonitrile (ACN)

The solubility of CP in ACN was also determined experimentally following a similar procedure

with minor modifications as described in Section 3.2.6.3. The procedure used when determining

solubility was conducted at room temperature (22˚C) rather than at 32˚C, and samples were

shaken at 120 rpm using a Model-3521 Junior Orbit Shaker (Lab-Line Instruments, Inc., Melrose

Park, IL, USA). The solubility of CP in ACN was found to be 491.12 ± 0.15 mg/ml (n = 3).

4

1.3.2. Dissociation constant (pKa)

CP does not have any ionisable functional groups and thus does not dissociate and therefore

there is no reported dissociation constant [20].

1.3.3. Partition coefficient

The octanol/water partition coefficient (log Po/w) of CP has been reported as 3.50 [21, 24]. The

log Po/w is defined as the logarithm of the partitioning ratio of a substance between octanol and

water, and is commonly used to quantitatively characterize the lipophilic nature of organic

compounds [25, 26]. Similarly lipophilicity has been described as a molecular parameter, which

may be used to describe the distribution equilibrium of a drug molecule between water and

various water immiscible, lipid-like organic solvents or other solubilising media such as, for

example, biological membranes [25, 27, 28].

Roberts et al., [29] argued that the lipophilicity of a solute is the main determinant for solute

partitioning into the stratum corneum from aqueous systems. Based on the log Po/w parameters,

CP can therefore be considered more lipophilic than hydrocortisone, which has a log Po/w of 1.61

[25, 29], and will more than likely partition into the stratum corneum from aqueous based semi-

solid formulations such as gels and creams faster than hydrocortisone would.

1.3.4. Melting range

CP has a melting range of approximately 195.5-197.0○C, at which temperature CP also

decomposes [21-23].

1.3.5. Optical rotation

The specific optical rotation of CP in a 1% w/v solution in 1,4-dioxan is +96º to +104º,

calculated with reference to a dry reference standard [22].

5

1.3.6. Stability

CP is unstable in the solid state and must be protected from light [22]. A solution or lotion of CP

should be stored at temperatures of between 4-25○C, and should not be used near an open flame

[14]. CP creams or ointments should be stored at a temperatures of between 5-30○C and the

cream should not be refrigerated [14, 15].

CP gel formulations should be stored at temperatures of between 2-30○C [14, 16] and CP foams

should be stored at controlled room temperatures of between 20-25○C and should not be exposed

to heat or stored at temperatures exceeding 49○C [14, 17]. Since the contents of a foam product

are usually under pressure, the container should also not be punctured, used or stored near heat or

an open flame, or placed into a fire or incinerator for disposal [14].

1.3.7. Ultraviolet absorption spectrum

The ultraviolet (UV) absorption spectrum of CP, determined experimentally in a binary mixture

of ACN:water (50:50) is depicted in Figure 1.2. The UV absorption spectrum was generated

using a Model-GBC 916 UV-VIS Double Beam Spectrophotometer (GBC Scientific Equipment

Pty. Ltd, Melbourne, Victoria, Australia), with the scanning range and speed set at between 200-

800 nm and 600 nm/min, respectively.

The data revealed that CP has a wavelength of maximum absorption (λ-max) of 240 nm. Despite

the use of different solvent systems, the absorption spectrum of CP obtained in these studies

using ACN:water (50:50) as a vehicle was similar to that previously reported in the literature

using methanol as the solvent or vehicle [21].

6

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0190 200 210 220 230 240 250 260 270 280 290 300

Wavelength (nm)

Ab

so

rba

nce

(mA

U)

240.0

Figure 1.2. UV absorption spectrum of CP in ACN:water (50:50)

1.3.8. Infra-red Absorption Spectrum

The infra-red (IR) absorption spectrum of CP was determined using potassium bromide (KBr)

pressed disks and the resultant spectrum is shown in Figure 1.3 [22]. The IR absorption spectrum

of CP shows principal peaks at wave-numbers 1666, 1612, 1724, 1063 and 1010 cm-1 [20] and

the relevant band assignments determined using theoretical concepts [30] are summarized in

Table 1.2.

7

Wavenumber (cm )-1

Tra

nsm

itta

nce

(%)

100

80

60

40

20

0

2000 1800 1600 1400 1200 1000 800 600 400

Figure 1.3. IR absorption spectrum of CP (adapted from 22)

Table 1.2. Major infra-red band assignment for CP

Band position (cm-1) Assignment 1666 C=C stretching of the aliphatic non-conjugated alkene 1612 C=O stretching of the ketone 1724 C-Cl stretching of chlorine 1063 COO stretching of the ether 1010 O-H bending of the alcohol

1.3.9. Crystal Structure

Haramura et al., [11] determined the crystal structure of CP using x-ray analysis and single

crystals of CP, purified by recrystallization from a methanol-acetonitrile solution. They reported

that the CP crystals belong to the group P21 and that the cell dimensions are a = 7.6961(3) Ǻ, b =

14.6036(5) Ǻ, c = 10.4355(5) Ǻ, β = 95.739(2) Ǻ and that the final discrepancy factor, R, is

0.038. Furthermore Harumara et al., [11] postulated that in the pregna-1,4-diene-3-one skeleton,

rings A, B, C (Figure 1.1) adopt planar, chair and chair conformations, respectively, whereas ring

D takes on an envelope shape at C14.

Haramura et al., [11] also argued that the stereochemistry of the methyl group at C16 is in the

alpha (α) configuration, which may result in steric hindrance between this methyl group and the

8

methyl group at C13. Nevertheless, they also postulated that the non-bonded C18 to C23 distance

of 4.103 Ǻ is indicative of the fact that such steric hindrance may be relieved.

According to Haramura et al., [11] the distance between the C20, carbonyl carbon and the O25

carbonyl oxygen is 2.742 Ǻ. The torsion angle, C17-O24-C25-O25 of -7.2º reveals that this part

of the molecule deviates significantly from the plane. The non-polarity of the CP is presumably

due to the C20-O25 interaction [11].

Haramura et al., [11] also reported that there are seven intermolecular interactions within the CP

molecule for which the geometrical parameters indicate that the interactions are due to hydrogen

bonds. Moreover Haramura et al., [11] argued that there is a single intermolecular oxygen bond

between the hydroxyl group and the carbonyl group attached to ring A.

1.4. STEREOCHEMISTRY AND STRUCTURE ACTIVITY RELATIONSHIP

All topical corticosteroids have a basic skeletal structure consisting of a fully reduced

phenanthrene ring system fused to a five-membered ring, giving rise to a

cyclopentanoperhydrophenanthrene nucleus, which is comprised of three six-membered and one

five-membered rings [31]. A further common attribute of the corticosteroid class of drugs is the

presence of methyl functional groups attached at positions 10, 13, 18 and 19 respectively (Figure

1.1).

The four rings of the corticosteroid skeleton do not exist in a flat place and the structure has no

elements of symmetry with each of the 19 positions being chemically distinct from each of the

other positions in the structure [31]. Furthermore, the corticosteroid skeleton is a rigid structure,

and it has been suggested that small changes in the position of a substituent usually results in a

significant change in the biological activity of the molecule [31].

Over the years, research has focused on strategies to optimise the potency and, in particular, the

anti-inflammatory and immunosuppressive capacity of the corticosteroids, while minimizing

adverse events and unwanted side effects associated with their use [9]. Although modifications to

9

the basic structure have led to the synthesis of topical corticosteroids with greater potency, these

new highly potent molecules are often associated with a greater potential to precipitate adverse

reactions to the molecule on administration [9].

As with all corticosteroids, CP consists of the basic four-ring 21-carbon structure. The 4,5 double

bond in ring A in addition to the 3-ketone group are essential for anti-inflammatory activity,

while the presence of a 1,2 double bond increases the glucocorticoid activity relative to

mineralocorticoid effects [9, 18, 31, 32]. The 9α-fluoro group in ring B enhances both

glucocorticoid and mineralocorticoid activity whereas the 11-hydroxy group in ring C is essential

for anti-inflammatory and glucocorticoid activity but not for mineralocorticoid effects [9, 18,

32].

The addition of a methyl functional group at position C16 in ring D eliminates mineralocortucoid

activity, while the presence of a propionate ester at position C17 and a chlorine atom at position

C21 increases topical activity due to a considerable increase in the lipophilicity of the molecule

[9, 18, 32]

1.5. CLINICAL PHARMACOLOGY

1.5.1. Mode of action

1.5.1.1. Overview

The physiological activity of corticosteroids can be divided into two broad categories, viz.,

mineralocorticoid effects that control electrolyte and fluid balance in the body and glucocorticoid

effects that influence carbohydrate, fat and protein metabolism [18, 33]. Although compounds

that lack the unwanted mineralocorticoid activity have been produced, to date it has proved

impossible to dissociate the glucocorticoid properties of synthetic topical corticosteroids from

their anti-inflammatory activity [33].

10

Topical corticosteroids are generally believed to have anti-inflammatory, antiproliferative,

antipuritic, and vasoconstrive properties [34]. However, it is through their anti-inflammatory

action that these agents are thought to exert their main therapeutic effects in a wide range of

acute and chronic inflammatory diseases, including most steroid-responsive dermatoses,

irrespective of aetiology [35].

1.5.1.2. Anti-inflammatory action

The precise mechanism by which CP and other topical corticosteroids exert their anti-

inflammatory effect in the treatment of steroid-responsive dermatoses is still uncertain [11, 15,

36]. However it is generally believed that at a cellular level, corticosteroids bind to specific

glucocorticoid receptors (GR) that are 777 amino acid protein members of the superfamily of

ligand-regulated nuclear receptors [9]. The GR have a modular structure and their principle

functions of transactivation, DNA binding and ligand binding are localised to specific domains in

that structure [36]. The GR are located and maintained in the cytoplasm as inactive multi-protein

complexes of two heat shock proteins (hsp90), which promote glucocorticoid binding but

prevent binding of the GR to DNA [9, 36].

The binding of a steroid to the GR is followed by dissociation of the hsp90 after which the

glucocorticoid-GR complex migrates into the nucleus of the cell and binds to DNA at specific

regions known as the glucocorticoid response elements on certain genes [34]. Successful binding

results in an increase in the production of lipocortin-1 [36]. Lipocortin-1 is a protein that has

been reported to belong to the annexin superfamily [9, 36], and its main function is to inhibit the

activity of phospholipase A2 directly, thereby decreasing the production of pro-inflammatory

prostaglandins, leucotrines, thromboxanes and leucocyte migration [36].

Apart from the direct regulatory effect on gene transcription, it has also been reported that topical

corticosteroids can regulate transcription of other transcription factors indirectly [34]. In

particular, corticosteroids have been shown to increase cellular levels of an inhibitory nuclear

factor (IκBα) by stimulating the expression of the IκBα gene [37]. IκBα then diffuses into the

cytosol and binds to the nuclear factor-κ B (NF-κB) thereby preventing translocation of NF-κB

11

to the nucleus and suppression of various gene products regulated by NF-κB, such as the

cytokines and adhesion molecules [37, 38]. Since NF-κB activates various immuno-regulatory

genes in response to pro-inflammatory stimuli, the inhibition of its activity can be a major

contributing factor in the anti-inflammatory activity of the glucocorticoids [38]. In this way,

corticosteroids may affect the transcription of genes that do not contain a glucocorticoid-

responsive receptor [34].

Topical corticosteroids have also been shown to inhibit the transcription of various pro-

inflammatory cytokine genes involved in skin disease such as interleukin (IL)-1, IL-2, IL-6,

interferon gamma (IFN-γ), and tumour necrosis factor-alpha [34]. Furthermore, corticosteroids

appear to stimulate lymphocyte expression of genes for anti-inflammatory cytokines such as

transforming growth factor-β and IL-10 [34]. Through regulation of cytokine production, it has

been suggested that corticosteroids probably play a role in rebalancing the T-helper cell type 1

(TH1) to TH2 lymphocyte ratio in skin lesions [34]. The anti-inflammatory effects that have been

associated with corticosteroid treatment include inhibition of capillary dilation and dermal

oedema and the suppression of endothelial cell and lymphocyte function [34].

1.5.1.3. Immuno-suppressive effects

Corticosteroids have also been reported to inhibit the proliferation of various cell types, such as

the T-lymphocytes [34]. However the anti-proliferative effects have not yet been clearly

defined, although it has been suggested that the effects are more than likely a consequence of the

blockade of cytokine expression and the suppression of cytokine effects [34]. Furthermore it has

been reported that some of the anti-proliferative effects of the corticosteroids may be mediated

through lipocortins which act as secondary messengers for corticosteroid compounds [34]. IL-10

and transforming growth factor-β1 have been shown to potentiate the inhibitory effects of the

corticosteroids on T-lymphocyte proliferation [34].

12

1.5.1.4. Anti-mitotic and vasoconstrictive effects

It has also been reported that topical corticosteroids have anti-mitotic and vasoconstrictive

activities in addition to their anti-inflammatory and immuno-suppressive actions [9]. The anti-

mitotic effects of the corticosteroids are secondary to the general reduction of protein synthesis

and may explain the therapeutic action of the corticosteroids in scaling dermatoses such as

psoriasis [9]. Similarly, the vasoconstrictive activity of the corticosteroids as demonstrated in

vascular beds may contribute to the anti-inflammatory activity of topical corticosteroids.

Although the mechanism by which vasoconstriction is induced is not yet completely elucidated,

it is thought to be related to the inhibition of natural vasodilators such as histamine, bradykinins

and prostaglandins [9].

1.5.2. Indications

CP is usually indicated in the management of significant inflammation of the skin that is

naturally thick or thickens as a result of disease, to such an extent that penetration of a less potent

topical corticosteroid would be poor, thereby making treatment difficult or ineffective [39].

Typical indications for CP use include the treatment of psoriasis of the body, palmoplanter

psoriasis, lichen planus, lichen simplex chronicus, lupus erythematosus, and acute exacerbations

of atopic dermatitis in adults [39].

CP is also indicated in the management of severe and acute attacks of any type of eczema and

chronic eczema, especially of the hands and feet where hyperkeratosis becomes an issue, chronic

hyperkeratotic sporiasis of the hands and feet, localised bullous disorders, keloids, pretibial

myoedema, vitiligo and also in the suppression of reaction after cryotherapy [20]. CP is

occasionally used for the management of light or photo-sensitivity reactions [20].

1.5.3. Contra-indications

CP preparations are contraindicated in patients with a known history of hypersensitivity to CP

and other related compounds or any excipients that may be used in the respective formulations

13

[15-17]. The use of CP in neonates and paediatric patients under the age of 12 is contraindicated

due to the potent local and more importantly systemic effects should they be absorbed [1, 10, 15,

20]. CP formulations should not be used in the treatment of acne vulgaris, rosacea or perioral

dermatitis or as monotherapy in the treatment of widespread plaque psoriasis [10, 14, 20].

CP should also not be used for the treatment of cutaneous infections caused by viruses, bacteria

or fungi [20]. If concomitant skin infections develop during CP therapy, appropriate antifungal

or antibacterial therapy must be initiated [14] and should the infection not respond to therapy, CP

should be discontinued until the infection is adequately controlled [14]. Skin infestations such as

scabies must not be managed with CP as exacerbations of the disease may occur or the disease

may be disguised [20]. The use of CP on the face, groin, or axillae is also contraindicated [20].

1.5.4. Adverse Effects

1.5.4.1. Overview

Since the commercial introduction of CP in 1973, primarily for the treatment of specific

dermatoses (Section 1.5.2), indications for the prescription of this potent corticosteroid have

been modified due to potential deleterious side effects that have been reported [40]. Serious side

effects nearly always follow the dispensing of uncontrolled repeat prescriptions [41] and have

been reported more frequently for the treatment of psoriasis than in eczema, probably due to the

fact that the parakeratotic keratin of psoriatic skin is more permeable than normal keratin [20].

The adverse effects are noticeable as localised skin reactions occurring at the site of application

and generalised adverse effects arising from systemic absorption of the compound [2, 9].

1.5.4.2. Local side effects

Burning, stinging, irritation and itching sensations have been reported as the most frequent local

cutaneous reactions in 1% of patients treated with CP creams in controlled clinical trials [15].

Less frequent local side effects include, cracking of the skin, erythema, folliculitis, numbness of

fingers, skin atrophy and telangiectasia [15]. The use of CP may exacerbate pre-existing or

14

coexistent dermatoses, such as rosacea, perioral dermatitis, tinea infections and resistant psoriatic

lesions [15].

1.5.4.3. Systemic side effects

Systemic side effects are usually not common but can arise when locally applied CP preparations

are absorbed through the skin and enter the general circulatory system [2]. The occurrence of

systemic absorption is rare and side effects are often clinically insignificant, particularly if the

formulation is used on an ad hoc basis [2]. The greatest risk of systemic side effects occurs when

large amounts of CP formulations are used over very large areas of the body for prolonged

periods of time [2, 20].

CP has the potential to cause suppression of the hypothalamic-pituitary (HPA) axis, Cushing’s

syndrome, diabetes and hypertension [20]. Suppression of the HPA-axis has occurred following

administration of topical dosages of as low as 2 g of the 0.05% w/w cream, ointment and gel or 7

g of the 0.05% w/w foam on a daily basis [15-17]. Cushing’s syndrome has been reported in

infants and adults following the prolonged use of topical CP formulations [15].

HPA-axis suppression, Cushing’s syndrome, linear growth retardation, delayed weight gain and

inter-cranial hypertension have also been reported in children being treated with topical

corticosteroids [15]. Manifestations of adrenal suppression in children include low plasma levels

and an absence of response to adrenocorticotropic hormone (ACTH) stimulation, whereas those

of intracranial hypertension include bulging fontanelles, headaches and bilateral papilledema

[15-17].

1.5.5. High risk groups

1.5.5.1. Pregnancy

The teratogenic potential of CP in humans is still unknown [14], however reproductive studies in

mice and rabbits using subcutaneous dosages of CP of up to 1 mg/kg and 10 µg/kg, respectively

15

revealed substantial harm, including cleft palate, skeletal immaturity, increased rates of still birth

and fetal resorption [14]. The administered doses were approximately 1.4 and 0.05 times those

that would be typical of topical doses in humans of either the CP cream or ointment formulations

[15-17].

Since CP can undergo percutaneous absorption [14] the data obtained from reproductive studies

suggest that the use of CP in pregnancy should be avoided [14]. It is also worth noting that

although no study has shown the absolute teratogenicity of CP, other potent corticosteroids have

been shown to be teratogenic in animals following topical application [14]. Therefore the use of

CP in pregnancy is contraindicated.

1.5.5.2. Lactation

Corticosteroids that are administered systemically are reportedly secreted in human milk and

could, therefore, suppress growth and interfere with endogenous corticosteroid production in

breast feeding infants [15]. It is unknown whether topical application of corticosteroids may

result in sufficiently high systemic levels following absorption to produce detectable quantities in

human milk. However since many drugs are excreted in breast milk, caution should be exercised

when CP cream, ointment, gel or foam formulations are administered to nursing mothers [15-17].

1.5.5.3. Paediatric use

The safety and effectiveness of CP cream, ointment, gel and foam formulations in paediatric

patients has not yet been established [15]. Consequently the use of CP in patients under the age

of 12 years of age is not recommended. Paediatric patients are at a greater risk of HPA-axis

suppression and Cushing’s syndrome than adults when they are treated with topical

corticosteroids [15]. This is a result of the fact that paediatric patients have a higher skin surface

area to body mass ratio than adults [15-17].

16

1.5.5.4. Geriatric use

In clinical studies in which CP cream, ointment, gel and foam formulations were used, a

sufficient number of patients over the age of 65 years were not included in the study to

categorically conclude whether geriatric patients respond differently to topical steroid therapy

than younger patients [15-17].

While other clinical experience has not revealed age-related differences in response, dosages

should generally be titrated carefully in geriatric patients by initiating therapy at the low end of

the dose range [15]. The greater frequency of decreased hepatic, renal or cardiac function and

concomitant disease drug therapy suggest that care must observed when treating elderly patients

with topical corticosteroid products [14]

1.6. PHARMACOKINETICS

1.6.1. Dosage and Administration

CP cream, ointment and gel should be applied sparingly as thin films and should be rubbed

gently into the affected area twice daily, preferably in the morning and evening [14, 15]. CP

foams and solutions should be applied to the affected areas of the scalp twice daily, in the

morning and evening [14, 17]. Some patients may respond initially to once daily or intermittent

therapy, for example, twice daily for three days per week [14].

The use of CP should be discontinued and a less potent topical corticosteroid preparation

substituted as soon as it is clinically feasible to alter therapy [10]. CP dosage should not exceed

50g of a CP 0.05% w/w cream, ointment, gel and foam or 50 ml of CP 0.05% w/v lotion per

week and the extended duration of a course of CP therapy should not exceed 14 days [10]. It has

been reported that many clinicians have indicated that prolonged CP therapy may be necessary in

rare cases in patients with resistant dermatological conditions, and careful monitoring of their use

in these patients is essential [10, 14].

17

1.6.2. Absorption

Percutaneous penetration and absorption of CP varies between individuals and can be altered by

using a variety of vehicles [14] and/or percutaneous penetration enhancers [42]. Absorption can

be increased by use of occlusive dressings [14, 41, 43] and use of percutaneous penetration

enhancers such as for example surface active agents, cyclodextrins or pyrrolidones [44, 45].

Following topical application of normal doses of CP to most areas of healthy skin, only small

amounts of the drug reach the dermis and subsequently the systemic circulation [10].

Nevertheless, systemic absorption may be increased when the usual dosage is exceeded or when

the skin is inflamed or diseased [10].

Mean peak plasma concentrations of CP of 0.63 ng/mL have been reported in one study, eight

(8) hours following a second dose of 30 g of CP that was applied thirteen (13) hours after an

initial dose of the 0.05% w/w CP ointment in healthy individuals with healthy skin. [14]. Mean

peak plasma concentrations of CP were slightly higher and occurred ten (10) hours after a second

dose of CP when a 0.05% w/w CP cream was administered to subjects [14]. Mean peak plasma

concentrations of approximately 2.3 and 4.6 ng/mL respectively have been reported to occur

approximately three (3) hours after a single application of a 25 g dose of a 0.05% w/w ointment

in patients with psoriasis or eczema, respectively [14].

1.6.3. Distribution

Advances in vehicle technology have enhanced the distribution of topical corticosteroids within

the skin structure [39] as illustrated by the results of an in vitro trial comparing the amount of

drug recovered after the application of 0.05% w/w CP lotion, cream or emollient cream [39].

Human skin samples sectioned by surgical excision were treated with different formulations and

the levels of CP in the skin were measured after centrifugation and extraction of the skin sample

[39]. It was found that significantly more CP was recovered in the epidermis, including the

stratum corneum, following administration of the lotion formulation as compared to that

following administration of a cream [39]. Similarly significantly more drug was recovered from

18

the epidermis and dermis following administration of the lotion as compared to that following

administration of an emollient cream [39].

1.6.4. Elimination

The absorption of CP into the systemic circulation seldom occurs (Section 1.6.2). Nevertheless

when percutaneous penetration and subsequent absorption of CP occurs, it has been reported that

the same metabolic pathways that are active when clearing systemically administered

corticosteroids are involved in the removal of the topically administered corticosteroid [14].

Although the systemic metabolism of CP has not been fully characterized or quantified [14], it

has been reported that the small amount of CP that may be absorbed following topical

application is metabolized in the liver [21] and excreted in the bile [14] and in the urine [14, 21].

1.7. CONCLUSIONS

The description, physicochemical characteristics, structural activity relationships, clinical

pharmacology and pharmacokinetics aspects of clobetasol 17-propionate (CP) have been

presented. CP is chemically known as 21-chloro-9α-fluoro-11β, 17α-dihydroxy-16β-

methylpregna-1,4-diene-3,20-dione 17-propionate. The empirical formula of CP is C25H32ClFO5,

and its molecular weight is 467.0 g/mole. CP is a synthetic di-halogenated analogue of

prednisolone and occurs as a white or almost white or cream-coloured crystalline powder which

is odourless.

Based on the human skin blanching assay CP has been labelled as a super-potent or class I

topical corticosteroid and is one of the most potent topical corticosteroids currently available on

the market. CP is marketed in a variety of vehicles such as creams, ointments, gels, lotions and

foams each containing 0.05% w/w of the corticosteroid.

CP does not have any ionisable functional groups and thus does not dissociate. The lack of such

functional groups results in CP having a relatively high octanol/water partition coefficient or log

Po/w of 3.5, making it relatively more hydrophobic than other corticosteroids such as for example

19

hydrocortisone. CP is therefore practically insoluble in water, but soluble in water-miscible

solvents such as acetonitrile or ethanol and freely soluble in some water-immiscible solvents

such as acetone, chloroform, and dichloromethane. The hydrophobic nature of CP suggests that

CP will partition into the stratum corneum from aqueous based topical formulations such as

creams and gels at a much faster rate than hydrocortisone.

The melting range of CP is reported to be approximately 195.5-197.0○C at which temperature it

also decomposes, while its specific optical rotation in a 1% w/v solution in 1,4-dioxan is between

+96º to +104º. The ultraviolet (UV) absorption spectrum of CP reveals that CP has a wavelength

of maximum absorption (λ-max) of 240 nm.

CP is unstable in the solid state and must be protected from light. Solutions and semi-solid

formulations containing CP should be stored at suitable temperatures of between 4-25○C for

solutions, 5-30○C for cream and ointments, 2-30○C for gels and 20-25○C for foams.

CP has structural features that are common to all topical corticortisteroids including 3-keto and

11-hydroxy groups, which are essential for anti-inflammatory activity, a 1,2 double bond and a

16-methyl group, which increases its glucocorticoid activity relative to the unwanted

mineralocorticoid effects. Functional groups that are unique to CP include a 9α-fluoro group that

enhances glucocorticoid and mineralocorticoid activity and a 17-propionate ester and a 21-

chlorine group, which increases the lipophilicity of the drug and hence its topical activity.

The mechanism of action of CP may involve anti-inflammatory, immuno-suppressive, anti-

mitotic and vasoconstrictive effects. The anti-inflammatory action of CP may be due to its

binding to specific glucocorticoid receptors (GR), which through a cascade of events decreases

the production of pro-inflammatory prostaglandins, leucotrines and thromboxanes and leucocyte

migration. The immunosuppressive effects of CP more than likely involve the blockade of

cytokine expression and the suppression of cytokine effects. Whereas the anti-mitotic effects of

CP may be secondary to a general reduction of protein synthesis the vasoconstrictive effects may

be due to the inhibition of natural vasodilators such as histamine, bradykinins and/or

prostaglandins.

20

CP is indicated for the treatment of significant inflammation of skin that is naturally thick or

thickens as a result of disease and where penetration of a less potent topical corticosteroid would

likely be poor. Some of the classical skin diseases that may require treatment with semi-solid

formulations of CP include psoriasis of the body, palmoplanter psoriasis, lichen planus, lichen

simplex chronicus, lupus erythematosus, and acute exacerbations of atopic dermatitis in adults.

Preparations containing CP are contraindicated in neonates and paediatric patients under the age

of 12 and in pregnant women, and caution should be exercised when CP formulations are

administered in nursing mothers as well as geriatric patients. As with most potent topical

corticosteroids CP may be associated with local and very seldom systemic side effects. The most

frequent local cutaneous reactions include burning, stinging, irritation and itching sensations and

although systemic side effects are very rare, they may include the suppression of the

hypothalamic-pituitary (HPA) axis, Cushing’s syndrome, diabetes and hypertension.

CP semi-solid formulations should normally be applied sparingly as a thin film and rubbed

gently into the affected area twice daily preferably in the morning and evening. CP dosage

should not exceed 50g of CP 0.05% w/w cream, ointment, gel and foam or 50ml of CP 0.05%

w/v lotion per week and the extended duration of a course of CP therapy should not exceed 14

days. Percutaneous penetration of CP is very rare when CP formulations are used appropriately

as described above. However in the unlikely event of CP absorption into the circulatory system

CP is likely metabolised in the liver and excreted in the bile and urine.

It is therefore, evident that CP has physicochemical and pharmacological properties, such as a

relative high lipophilicty and local anti-inflammatory activity, respectively, which make the drug

suitable for incorporation into semi-solid formulations, such as creams, ointments, gels and

foams for topical administration. Consequently, CP was selected as an ideal candidate for

inclusion into a generic cream formulation developed and assessed in these studies.

21

CHAPTER TWO

DEVELOPMENT, OPTIMIZATION AND VALIDATION OF AN HPLC METHOD FOR

THE ANALYSIS OF CLOBETASOL 17-PROPIONATE IN SEMI-SOLID DOSAGE

FORMS

2.1. INTRODUCTION

The quantitative analysis of clobetasol 17-propionate (CP) in semi-solid formulations has

primarily been accomplished using reversed-phase high performance liquid chromatography

(RP-HPLC) with ultraviolet detection [3, 10, 46-48]. Several other analytical techniques that

have been used for the quantitation of CP in dosage forms include normal-phase HPLC,

ultraviolet spectroscopy, liquid chromatography-mass spectrometry (LC-MS) and liquid

chromatography-mass spectrometry-mass spectrometry (LC-MS-MS or LC-MSn) [47].

The Food and Drug Administration (FDA) has published a guidance document [49] in which it

recommends that an appropriate, specific and sensitive analytical procedure be used to analyse

and determine drug concentrations and the amount of drug released from dosage forms during in

vitro release studies [49]. It has been reported that a major difficulty encountered in the analysis

of semi-solid dosage forms is the potential interference due to formulation adjuvants and

preservatives that are usually present in what are relatively complex formulations [50].

Generally, only one or two components are required to be quantitated and these components

must be adequately separated from formulation exicipients, which may interfere with the assay

procedure [50].

RP-HPLC is a commonly used, powerful and reliable analytical tool that can be used for the in

vitro analysis of formulations such as creams, ointments and gels that are of a complex nature,

since HPLC not only provides separation and quantitative data but also has the ability to

eliminate almost all interference problems [51]. The objective of these studies was therefore to

develop, optimize and validate a simple, selective, sensitive, precise, accurate and linear RP-

22

HPLC method that is suitable for the quantitative analysis of CP in cream formulations and CP

release during in vitro release studies.

2.2. PRINCIPLES OF RP-HPLC

Liquid chromatography (LC) is a method of chromatographic separation based on the difference

in distribution of an analyte between two immiscible phases, in which the mobile phase is a

liquid and percolates through a stationary phase, usually contained in a column [52]. Although

various terms, including high-speed LC, high-efficiency LC and high-pressure LC, have been

used to describe LC, high-performance liquid chromatography (HPLC) is now the generally

accepted terminology [52].

In HPLC, a stationary phase is either coated onto a finely divided inert support or chemically

bonded to a support material, contained within a stainless steel tube over which the mobile phase

flows, thereby affecting the separation of individual components of a mixture [52, 53]. Coated

phase LC uses a bulk liquid stationary phase, which is mechanically held to the support by

adsorption [54], whereas in bonded phase chromatography (BPC), an organic stationary phase is

chemically bonded to a support material rather than being held in place by mechanical means

[55].

BPC packing materials have been reported to be more stable than coated phase materials due to

the fact that a stationary phase that is chemically bound to a support cannot be easily removed or

lost during normal use of the column [56]. However, the poor reproducibility of some packing

materials that are commercially available has been mentioned as being one of the major

disadvantages of using BPC [56]. Nevertheless, the success of BPC has virtually eliminated the

use of coated stationary phases in chromatography [52].

In HPLC, the sample to be analysed is dispersed into a mobile phase and the analyte(s) of

interest pass through the stationary phase by pumping the mobile phase through the stationary

phase using a solvent delivery module [53, 55]. As the analyte molecules pass through the

column, there is constant interaction between the solute molecules, the stationary phase and the

23

mobile phase [53, 55]. Separation of the various components of a mixture is reported to occur as

a result of differences in equilibrium of distribution of different solute molecules in the sample

undergoing analysis [53, 55].

Two principle modes of HPLC, viz., normal-phase HLPC (NP-HPLC) and reversed-phase HPLC

(RP-HPLC) can be distinguished by the relative differences in the nature of the stationary phases

used to effect a separation in addition to the corresponding mobile phase composition and

differences in the nature of the interaction of functional groups present in solute molecules with

these phases [52, 53]. The nature of the functional groups of a molecule dictate the selectivity

and specificity of an interaction between an analyte and the column support material or mobile

phase, leading to the selectivity and specificity of a separation [53].

NP-HPLC requires the use of a polar stationary phase and a non-polar mobile phase in order to

separate analyte(s) of interest [53, 57]. In NP-HPLC the stationary phase may be either an

adsorbent, such as silica, or a liquid phase that is bonded to a solid support. The mobile phase

may consist of either a single non-polar solvent, such as hexane, or a mixture of non-polar

solvents or a non-polar solvent mixed with a small amount of a polar non-aqueous solvent, such

as methanol [57].

NP-HPLC is most useful for the separation of compounds of moderate to strong polarity since

non-polar solutes elute near the mobile phase front [53]. NP-HPLC is normally restricted to the

separation of stereochemical isomers, diastereomers, low molecular weight aromatic compounds

and long chain aliphatic compounds [53].

In contrast, RP-HPLC entails the use of a hydrophobic bonded stationary phase with a mobile

phase that consists of polar solvents such as water, with or without buffers or mixtures of water

and water-miscible organic solvents such as methanol and acetonitrile [53, 57, 58]. RP-HPLC is

usually the first choice of method for use in most pharmaceutical applications and especially for

the analysis of neutral or non-polar compounds that dissolve in water-organic solvent mixtures

[59]. Since the majority of pharmaceutical compounds of interest are relatively non-polar, most

HPLC analyses in pharmaceutical research are carried out using RP-HPLC techniques [57].

24

BPC columns for both NP-HPLC and RP-HPLC may be categorized according to the length and

functionality of the organic side chain used to manufacture the column [53]. Typically carbon

chains ranging in length from three (3) to eighteen (18) carbon atoms, viz., C3, C4, C8 and C18,

are chemically bonded to a silica support surface [53]. If the organic side chain is only alkyl in

functionality, the resulting phase is hydrophobic and thus useful for the reversed -phase mode of

analysis [53]. Consequently, stationary phases for RP-HPLC are prepared by reaction of residual

silanol functional groups in a silica backbone with n-alkylchlorosilanes [58].

RP-HPLC is used extensively in various scientific fields viz., pharmaceutical, agricultural and

medical sciences, in addition to fundamental studies in the separation sciences [60]. However,

despite its extensive application, the retention mechanism of molecules in RP chromatographic

process has not yet been fully elucidated [56, 60]. This is more than likely due to the complexity

of RP-HPLC systems, the properties of which have been reported to change dynamically with

the composition of the mobile phase and type(s) of the stationary phase used in separations [61].

Nevertheless, some investigators have alluded to the fact that separation in RP-HPLC may be

due to either adsorption effects or partitioning of a solute between a stationary phase and a

mobile phase, or combinations thereof [53, 56].

2.3. METHOD DEVELOPMENT

2.3.1. Overview

A summary of RP-HPLC with UV detection methods that have been developed and used for the

quantitative determination of CP in semisolid dosage forms is listed in Table 2.1. The data

contained in these reports were used to set preliminary RP-HPLC conditions for the development

of a suitable in vitro analytical method for the quantitation of CP in cream formulations and for

the characterisation of release of CP from topical formulations developed in these studies.

25

Table 2.1. RP-HPLC methods used for the analysis of CP in semi-solid dosage forms

Column Mobile phase Flow rate (ml/min) Detection method Reference µBondapak® C18 300 mm x 3.9 i.d methanol:water (75:25% v/v) 1.5 UV-240 nm 46 Nova-Pak® C18 5-µm 150 mm x 3.9 mm i.d. methanol:water (70:30% v/v) 1.0 UV-254 nm 10 Lichrospher® C18 125 mm x 4.0 mm i.d. methanol:aqueous phase pH 3.5

adjusted with 0.05% citric acid (65:35% v/v)

1.1 UV-240 nm 3

Nova-Pak® C18 4-µm 150 mm x 3.9 mm i.d. acetonitrile:water (50:50% v/v)

1.0 UV-240 nm 47

Purospher-Lichrocart® 5-µm C18 250 mm x 4.0 mm i.d.

acetonitrile:water (40:60% v/v) 1.0 UV-237 nm 48

Phenomenex® Luna C18 150 mm x 4.6 mm i.d. methanol: 0.1M KH2PO4 pH 3: acetonitrile (10:40:50% v/v)

1.0 UV-239 nm 62

L1® 150 mm x 4.6 mm i.d. acetonitrile:0.05M NaH2PO4 pH 2.5: methanol (95:85:20% v/v)

1.0 UV-240 nm 23

Spherisorb ODS® 2, 5µm, 50 mm x 4.6 mm i.d. methanol:0.05M NaH2PO4 pH 2.5: acetonitrile (10:42.5:47.5% v/v)

1.0 UV-240 nm 22

26

2.3.2. Experimental

2.3.2.1. Chemicals

All chemicals used in these studies were at least of analytical reagent grade. HPLC-grade

solvents, viz., methanol (UV cutoff of 215 nm) and acetonitrile (UV cutoff of 200 nm) were

purchased from Romil Ltd. (Waterbeach, Cambridge, UK). HPLC-grade water was prepared

using a Milli-RO® 15 water purification system (Millipore Co., Bedford, MA, USA) that

consisted of a Super-C® carbon cartridge, two Ion-X® ion-exchange cartridges and an Organex-

Q® cartridge. The water was filtered through a 0.22 μm Millipak® 40 stack filter (Millipore Co.,

Bedford, MA, USA) prior to use.

Clobetasol 17-propionate (CP) was purchased from Symbiotec Pharmalab P.V.T. Ltd

(Pigdamber, Maharastra, India). Betamethasone 17-valerate (BV) was purchased from Sigma-

Aldrich SA (Pty) Ltd (Brackenfell, Western Cape, RSA) and Dermovate® cream (Glaxo

Wellcome SA (Pty) Ltd, Midrand, Gauteng, RSA) was purchased from Wallaces Pharmacy

(Grahamstown, Eastern Cape, RSA).

2.3.2.2. Instrumentation

Two modular HLPC-UV chromatographic systems, i.e., System 1 and System 2, were used in

these studies. System 1 was used for the development, optimization and validation of the

analytical method and System 2 for the characterisation of CP content and release in topical

formulations during product development and assessment studies. A mini-revalidation of the

analytical method was undertaken prior to using System 2 for quantitative purposes.

System 1

The modular HPLC-UV system consisted of a Beckman Model 112 Solvent Delivery Module

(Beckman Instruments, Inc., San Ramon, CA, USA), a WISP™ Model 712 Autosampler

(Millipore® Waters Associates, Milford, MA, USA) and a linear UV-100 detector

27

(Spectrachrom, NV, USA) set at λ = 240 nm. Data acquisition was performed using a dual-pen

Model 561 strip chart recorder (Perkin-Elmer, Maywood, IL, USA) and separation was achieved

on a Nova-Pak® 60 Å C18 4-µm (3.9 i.d. x 150 mm) cartridge column (Millipore® Waters

Associates, Milford, MA, USA).

System 2

The modular HPLC-UV differed from System 1 in that it consisted of an Isochrom LC dual

piston solvent delivery module (Spectra-Physics, San Jose, CA, USA) and data acquisition was

performed using an SP-4600 Integrator (Spectra-Physics, San Jose, CA, USA). All the other

system components were the same as those used for System 1.

2.3.2.3. UV detection of CP and internal standard

The ideal detector for HPLC should be sensitive, respond to all solutes universally or have

predictable specificity, have a linear response over several orders of concentration, possess a low

dead volume, be non-destructive, be insensitive to changes in temperature and mobile phase

velocity changes, operate continuously and be reliable and convenient to use [52]. Unfortunately,

no single detector is able to satisfy all these criteria and consequently analytical methods have to

be optimized for specific pieces of equipment.

In most cases, analytical method development for HPLC commences by use of ultraviolet (UV)

detection with either a variable-wavelength spectrophotometric or diode-array detector [63]. UV

detector technology has developed over the years and is associated with good sensitivity and

linearity and can provide an adequate response for most samples, except when samples have

little or no UV absorbance, analyte concentrations are too low for UV detection, sample

interference is important, and/or qualitative structural information about a molecule is required

[63].

It has been suggested that the manner in which a UV detector is used may affect the relative

response of sample components and could potentially interfere with the selectivity, sensitivity

28

and baseline noise of an analytical method [63]. The achievement of adequate sensitivity is

primarily dependent on the selection of an appropriate wavelength for analysis, which is chosen

from knowledge of the UV absorption spectrum of the individual sample components to be

evaluated [63]. The UV absorption spectrum of an analyte should therefore be evaluated prior to

the development of a comprehensive HPLC method for analysis of that compound [63].

The analysis of CP by RP-HPLC has been accomplished using UV detection at 237 nm [64], 239

nm [62], 240 nm [3, 46, 47] and 254 nm [10]. An evaluation of the UV absorption spectrum of

CP (Figure 1.2, Section 1.3.7.) reveals that the wavelength of maximum absorption (λmax) is 240

nm. Consequently, the eluent in these studies was monitored at 240 nm. The detection of the

internal standard does not have to occur at the λmax of the compound in question, since enough

material may be added to a sample to achieve an adequate response from the molecule, at the

desired attenuation used for analysis of the compound of interest.

2.3.2.4. Column selection

It has been suggested that the column is the heart of any HPLC separation procedure [65, 66]. It

follows therefore that the availability of a stable, high-performance column is vital for the

development of a rugged, reproducible and reliable analytical method. The stability of a column

is especially important in method development since once a desired separation is achieved the

column characteristics should remain unchanged for as long as possible and excellent column

stability minimizes the need for further adjustment of the separation conditions or replacement of

a column during any specific application of the method [65].

Commercially available columns may differ in their reproducibility when purchased from

different suppliers and in many cases differences exist in columns supplied from a single supplier

[66]. Such differences can have a serious impact on developing a reproducible and useful HPLC

method of analysis for a particular compound [65]. The decision to use a particular HPLC

stationary phase is based on physicochemical characteristics such as the solubility, molecular

weight and ionic nature of the analyte of interest, in addition to the column packing material and

dimensions of the column [66, 67].

29

The majority of column materials used for HPLC separations make use of a silica particle base or

support, although there are a number of columns that are now available that are packed with

porous-polymeric support materials [65, 66, 68]. However, the majority of RP-HPLC separation

techniques use silica-based bonded-phase chromatographic (BPC) columns [58, 66]. Such

columns are made by reacting mono-functional chorodimethylsilanes with available silanol

functional groups of a stationary phase material [58, 65, 66].

A variety of different alkyl and substituted alkyl silica materials are formed as a consequence of

these reactions. Some of the examples of the types of bonded phase materials that have been

produced include n-octadecylsilane (ODS or C18), n-octylsilane (C8), phenyl, dimethylsilane and

dimethylamino functionalities [65, 66]. Separation of compounds on BPC columns in RP-HPLC,

primarily depend on molecular interactions that occur between a solute and the components of

the mobile and stationary phases being used for that separation [69]. The underlying

mechanisms of a separation include hydrophobic and solute-solvent interactions that result in

increased retention of compounds with large C-H or hydrophobic surface areas and shorter

retention times for compounds containing polar or hydrophilic functional groups such as

hydroxyl groups (OH) [68].

The retention process in RP-HPLC involves partitioning of solutes between a mobile phase and

the bonded alkyl groups of a stationary phase that contain organic solvents extracted from the

mobile phase in use [68]. The retention of an analyte may be affected by the composition of the

mobile phase and the structure of the stationary phase. Consequently, the length and surface

density of alkyl chains in a stationary phase have been reported to be the primary determinant of

the phase ratio between the mobile and stationary phases of a column, the orderliness of alkyl

chains in that column and the solvent content (or polarity) of the stationary phase [68] .

The stability of silica-based BPC columns depends on the types of silica support and bonded

phases used to manufacture the column, the mobile phase pH and the type of the buffer and

organic modifier(s) used to prepare the mobile phase. The loss of silane-bonded phases can occur

due to hydrolysis of the siloxane bonds (Si-O-Si) that bind the silane to the support backbone of

30

the column. Furthermore, degradation of supports is accentuated at high temperatures, low pH

and in highly aqueous mobile phases [65].

The most commonly used columns that have been reported for the analysis of CP in dosage

forms are octadecylsilyl (C18) based stationary phases[3, 10, 46-48, 70], which may be due to the

fact that CP is a highly hydrophobic drug with an octanol-water partition coefficient of 3.5 [47].

Therefore, retention and separation of CP on such columns can be easily accomplished using a

mixture of water and water-miscible solvents such as methanol and/or acetonitrile. Other

common features of these columns are the average particle size of the packing materials, which

are ≤ 5 µm, and the column dimensions, which are 150 mm (L) x 3.9-4.6 mm (i.d.) (Table 2.1,

Section 2.3.1). It has been suggested that the retention of a compound on an RP-HLPC analytical

column may be affected by the column packing materials as well as the column dimensions [65,

66].

It is generally believed that the efficiency of packed RP-HPLC columns is enhanced with a

decrease in the particle diameter of the stationary phase, such that the smaller the particles the

better the resolution and sensitivity that can be obtained with the column of interest [65, 66].

Most RP-HPLC columns are packed with particles with sizes ranging between 3-10 µm and

although smaller particles will generally result in high theoretical plate numbers (Section 2.3.2.5)

and better sensitivity, they more than likely lead to high column back-pressures than the packing

materials with large sizes [65, 66]. Consequently, RP-HPLC columns packed with particles with

an average diameter of 5 µm may represent a good compromise in terms of column efficiency,

back-pressure and lifetime of use [65].

As far as column dimensions are concerned, it has been suggested that columns of lengths of

between 100-1500 mm may be used for HPLC analysis, although most of the newer equipment

allows for the use of columns with maximum lengths of only 250 mm [66]. Kromidas [71]

suggested that the use of a longer column may improve resolution at the expense of retention

times and column back-pressures. Similarly, the efficiency of well-packed columns of small

particles of ≤ 10 µm in size has been reported to increase with an increase in the column internal

diameter (i.d.) [66]. However, HPLC columns with i.d. of between 4-5 mm have been reported to

31

represent a good compromise of performance, convenience, amount of mobile phase used and

column packing required in analytical applications [66].

A Nova-Pak® 4 µm (150 mm x 3.9 i.d.) cartridge column, packed with dimethyl octadecylsilyl

(C18) bonded amorphous silica was therefore selected for testing and thereafter for the

development of an analytical method for the determination of CP in dosage forms.

2.3.2.5. Column efficiency

The efficiency of a chromatographic column can be expressed quantitatively as the number of

theoretical plates (N) for a test substance conducted under favorable conditions and may be

calculated using equation 2.1 [72].

2

21

54.5 ⎟⎟⎠

⎞⎜⎜⎝

⎛=

WtN R Equation 2.1

Where,

tR = retention time of a test peak

W1/2 = peak width at half peak height.

The value obtained for N for a column increases as the mean diameter of the packing material

decreases, due to an increase in the associated surface area of the stationary phase available for

interaction with the mobile phase and the analyte of interest [57]. Particles with an average

diameter of 10 µm are considered standard for RP-HPLC [57].

Such material provides excellent efficiency with an estimated theoretical plate count of

approximately 8000 plates per column in addition to the ability to use relatively high solvent

flow rates without exceeding the pressure limits of a system [57]. Packing particles of 5 µm

diameter yield an approximate doubling of the column efficiency as compared to 10 µm packing

materials but yield substantial increases in system back pressures at given mobile phase flow

rates [57].

32

The efficiency of a column as measured by N is best determined with an ideal test system, rather

than with the analyte for which the method is being developed and the conditions used during

method development [72]. Column test systems may be comprised of a small, neutral test

compound such as toluene or naphthalene, a flow rate of 1 ml/min and a mobile phase with a

viscosity (η) of less than 1 cP, such as compositions of 0 to 100% acetonitrile-water mixtures

maintained at temperatures of less than 20˚C [72].

The efficiency of the column selected for use in these studies, a Nova-Pak® C18 4 µm (150 mm x

3.9 i.d.) was interrogated by injection (n = 6) of a test mixture containing uracil, acetophenone,

benzene, toluene and naphthalene onto the column at room temperature (22˚C). The separation

was achieved using a mobile phase composition of acetonitrile-water (65:35) at a flow rate of 1

mL/min. Detection of the compounds was conducted at 254 nm and a typical chromatogram of

the separation of the test mixture under these conditions is depicted in Figure 2.1.

33

1

2

3

4

5

6420

Retention time (min)

inje

ctio

n

Figure 2.1. Typical chromatogram of a test mixture containing uracil (1), acetophenone (2), benzene (3), toluene (4) and naphthalene (5) after separation on a 4 µm Nova-Pak® C18 (150 mm x 3.9 i.d.) cartridge column.

The calculation of the theoretical plate number gives an indication of the efficiency and

performance of a column. The Nova-Pak® C18 4 µm (150 mm x 3.9 i.d.) cartridge column yielded

an average efficiency of 6905 ± 777 theoretical plates (n = 6). Although ideally a column of such

small particle diameter (4 µm) and length (150 mm) should give an N value of more than 10000

[65], the column was considered efficient and suitable to provide reproducible separations of CP

while minimizing solvent usage at low flow rates, and it was therefore selected for method

development and validation studies. The low column efficiency may have been due to the fact

that the column was not new as it had been used for other applications in our laboratory.

34

2.3.2.6. Internal standard selection

An internal standard (IS) is a compound added in equal amounts to all standards and test samples

to be analyzed [73] and is used to improve the accuracy of an analytical method by

compensating for varying injection volumes and day-to-day instrumental changes [74, 75]. The

physicochemical and analytical properties of an ideal IS should be similar to those of the analyte

of interest [73]. Clobetasol butyrate [10], p-phenylphenol [3] and uracil [46] are some of the

compounds that have used as IS for the analysis of CP.

Potential internal standards that were tested included uracil (UR), prednisone (PR),

triamcinolone (TA), pregnenolone carbonitrile (PC), stilboestrol dipropionate (SD),

hydrocortisone acetate (HA), betamethasone 17-valerate (BV) and mometasone furoate (MF).

Separation was achieved using a mobile phase composition of acetonitrile-water (50:50) at a

flow rate of 1 mL/min. Detection of compounds was conducted at 254 nm and the data generated

in these studies are depicted in Table 2.2.

Table 2.2. Retention times of CP and potential internal standards

Compound Retention time Clobetasol 17-propionate (CP) 8.2 Uracil (UC) 1.0 Prednisone (PR) 1.0 Triamcinolone (TA) 1.2 pregnenolone carbonitrile (PC) 2.0 stilboestrol dipropionate (SD) 2.0 hydrocortisone acetate (HA) 2.2 betamethasone 17-valerate (BV) 5.4 mometasone furoate (MF) 8.8

The retention times of UR, PR, TA, PC, SD and HA were considered too short and therefore

undesirable. Consequently, UR, PR, TA, PC, SD and HA were not considered as suitable for use

as IS, since the short retention times could result in interference from the solvent front or any

other peaks that elute close to the solvent front during HPLC analysis. Similarly, MF was not

selected since it eluted close to the retention time of CP. Only BV was eluted at a suitable

retention time and at a distance from the solvent front without interfering with the CP peak.

Consequently BV was selected as the most suitable IS for the analysis of CP.

35

2.3.2.7. Mobile phase selection

It is worth noting that the development of a successful RP-HPLC separation does not rely solely

on the selection of a suitable column, but also on matching an appropriate mobile phase to a

specific column, analyte and sample matrix [76]. The retention times of compounds in RP-HPLC

are adjusted by changing the mobile phase composition or solvent strength, which in turn

depends on the choice of organic solvent and the concentration of that solvent in the mobile

phase [67].

Conversely, the correct choice of organic solvent requires a knowledge of the physicochemical

characteristics such as boiling point, viscosity and UV absorbance of that solvent [76, 77].

Furthermore, solvent purity, cost and the impact of the solvent on retention of the analyte, in

addition to interactions with the analyte and the stationary phase, should be carefully considered

when selecting an organic modifier for use in RP-HPLC analysis [76-78].

Methanol (MeOH) and acetonitrile (ACN) are the most commonly used organic modifiers in RP-

HPLC [79]. Tan et al., [80] reported that in hydro-organic mixtures, ACN and MeOH may self-

associate or associate with water molecules to form clusters, albeit to different extents. As the

polarity of methanol is greater than that of acetonitrile, the former forms hydrogen bonds by

accepting or donating protons, whereas the latter being aprotic is unlikely to form hydrogen

bonds [80]. Therefore the difference in hydrogen bonding ability may influence the adsorption of

a modifier into a stationary phase as well as solute partitioning, thereby influencing a resultant

separation [80].

CP is a hydrophobic compound and is practically insoluble in aqueous solutions, whereas it is

freely soluble in water-miscible solvents, such as acetone, chloroform and dichloromethane

(Table 1.1, Section 1.3.1.1). HPLC analysis of CP has therefore been achieved using bonded

phase columns (C18) and mobile phases containing either MeOH or ACN as the organic modifier

of choice (Table 2.1). The use of MeOH and water as the mobile phase in this analytical method

resulted in asymmetric peak shapes with a high degree of peak tailing and exceedingly high

column back-pressures. In contrast, excellent peak symmetry, resolution, a reduction in peak

36

tailing and column back-pressure were observed when mobile phases of binary mixtures of ACN

and water were used. Typical chromatograms generated using a binary mixture of MeOH and

water and a binary mixture of ACN and water are shown in Figure 2.2.

121086420


inje

ction

121086420


inje

ctio

n

A

B

Figure 2.2. Typical chromatograms generated using a binary mixture of MeOH and water (A) and a binary mixture of ACN and water (B) as mobile phase (chromatograms graphically redrawn).

ACN-water mixtures have lower viscosities than the corresponding MeOH-water mobile phases,

resulting in higher theoretical plate numbers and lower column back-pressures, which ultimately

result in better peak shapes [67, 79]. Although MeOH is less expensive and apparently less

damaging to the environment than ACN [79] the use of MeOH-water binary mixtures as mobile

37

phase for this analytical method was considered inappropriate. In order to prolong the lifespan of

the column and develop an effective separation for the quantitation of CP in semi-solid dosage

forms ACN was used as the organic modifier and was the only solvent evaluated in these studies.

2.3.2.8. Preparation of mobile phase

Mobile phases were prepared by adding equal parts by volume of HPLC-grade ACN and HPLC-

grade water that had been prepared using a Milli-RO® 15 water purification system (Millipore

Co., Bedford, MA, USA) to a glass Duran® Schott solvent mixing bottle (Schott Duran GmbH,

Hattenbergstrasse, Germany). The mixture was allowed to equilibrate to room temperature and

the mobile phase was filtered through a 0.45 µm Millipore® HVLP filter (Millipore, Bedford,

MA, USA) and degassed under vacuum with the aid of a Model A-2S Eyela Aspirator (Rikakikai

Co., Ltd, Tokyo, Japan) prior to use. Degassing of mobile phases is necessary to remove

dissolved oxygen, which could lead to the formation of air bubbles in the flow cell of a detector

or in the connecting tubing, thereby affecting the reproducibility and sensitivity of the detection

system and the flow rate [77]. The mobile phase was freshly prepared daily and was not recycled

during use.

2.3.2.9. Preparation of stock solutions and calibration standards

Standard stock solutions of CP (0.1 mg/ml) and BV (0.5 mg/ml) were prepared by accurately

weighing 10 mg of CP and BV using a Model AG-135 Mettler Toledo top-loading analytical

balance (Mettler Instruments, Zurich, Switzerland) into 100 ml and 20 ml A-grade volumetric

flasks, and dissolving in 20 ml and 5 ml ACN, respectively. The stock solutions were placed in a

Model 8845-30 ultrasonic bath (Cole-Parmer Instrument Comp. Chicago, IL, USA) for 5 min, in

order to ensure complete dissolution of the drug, after which samples were made up to volume

with ACN. Stock solutions were protected from light using aluminium foil and were, stored in a

refrigerator at 4˚C. Stock solutions were used within a maximum period of two weeks based on

stability study data generated as described in Section 2.4.7.3. Calibration standards of CP were

prepared by serial dilution of the stock standard solution on a daily basis to produce solutions of

concentration of 0.10, 0.50, 1.0, 3.0, 6.0, 12 and 18 µg/ml, using mobile phase as the solvent.

38

2.3.2.10. Effect of ACN concentration

The impact of ACN concentration on the retention times of CP and BV was evaluated using a

variety of binary mixtures of ACN and water (ACN/water). The mixtures varied in composition

from 40–65% ACN and the data generated in these studies are depicted in Figure 2.3.

0

2

4

6

8

10

12

35 40 45 50 55 60 65 70

ACN concentration (% v/v)

Ret

entio

n tim

e (m

in)

BV CP

Figure 2.3. Effect of ACN concentration on the retention times of BV and CP

It is clearly evident that BV has a shorter retention time (Rt) than CP with any of the ACN/water

binary mixtures tested. This is more than likely due to the fact that CP by virtue of its

physicochemical characteristics is relatively more hydrophobic than BV and therefore interacts

to a greater extent with the stationary phase than BV does.

The data also show that the retention times of BV and CP were inversely proportional to the

ACN concentration in the mobile phase. In other words an increase in the ACN content of the

mobile phase composition led to shorter retention times for both CP and BV. For example, at a

39

low ACN concentration of 40% v/v, BV had an Rt of 9.00 min, whereas CP had an Rt of 11.20

min. However, at a high ACN concentration of 60% v/v, BV had an Rt of 2.45, whereas CP had

an Rt of 3.28 min. The decrease in Rt for both BV and CP with an increase in the ACN content is

more than likely due to enhanced solute-solvent interactions and diminished solute-stationary

phase interactions.

The aim of conducting these studies was to determine an optimal ACN concentration for use in

the mobile phase that would produce acceptable retention times for both CP and BV. In our

laboratory, retention times are considered acceptable when the first peak of interest, which is

either the IS or the API, elutes 4 min after the solvent front, with the second peak of either the IS

or the API eluting 2-4 min later, resulting in a maximum run time of approximately 10 min.

Based on these criteria and the data illustrated in Figure 2.3 a binary mixture consisting of 50%

w/w ACN-water was selected as the most suitable mobile phase for the separation of CP and BV.

2.3.2.11. Effect of flow rate

The effects of mobile phase flow rate on the retention times of CP and BV are illustrated in

Figure 2.4. The results of these studies indicate that, as expected, the retention time of both CP

and BV decrease as mobile phase flow rate was increased. A rapid HPLC separation with respect

to a run time of approximately 10 min for the analysis of CP in pharmaceutical formulations was

desired. A flow rate of less than 0.8 ml/min resulted in a CP retention time of greater than 10

min, which was considered unacceptable since the method would be time-consuming and result

in long overnight analyses.

40

0

2

4

6

8

10

12

14

0.5 0.7 0.9 1.1 1.3 1.5

Flow rate (ml/min)

Ret

entio

n tim

e (m

in)

BV CP

Figure 2.4. Effect of mobile phase (50 %v/v ACN) flow rate on the retention times of BV and CP

Although increasing the flow rate to 1.4 ml/min resulted in relatively short retention times for

both CP and BV, closer inspection of the chromatogram revealed some peak shouldering,

especially in the case of BV peak, which was possibly due to poor column performance at high

flow rates. However, when the flow rate was set at 1.0 ml/min, sharp well-resolved peaks with

the desired retention times for both BV and CP were observed (Figure 2.5). Consequently the

mobile phase flow rate was set to 1.0 ml/min and this rate was used for all subsequent analyses.

2.3.2.12. Optimal mobile phase composition and flow rate

The final mobile phase selected for the analysis of CP using BV as the IS was a binary mixture

of ACN and water in a ratio of 50:50 or 50 %v/v ACN. The peak shape and retention times were

found to be suitable when this mobile phase composition was used with excellent resolution

between CP and the IS. The mobile phase flow rate was set at 1.0 ml/min with resultant retention

times of 5.6 min and 8.2 min for BV and CP respectively.

41

The total run time for sample analysis was 10 min and the use of this flow rate and resultant run

time ensured that comparatively low volumes of mobile phase and solvents were consumed by

the analysis of CP in test samples. A typical chromatogram of the separation achieved using this

mobile phase and the reported chromatographic conditions, is shown in Figure 2.5.

121086420


CP

BV

inje

ctio

n

Figure 2.5. Typical chromatogram of a mixture of the internal standard, betamethasone 17-valerate (BV) and clobetasol 17-propionate (CP) using a mobile phase of 50% v/v ACN-water and a flow rate of 1.0 ml/min

42

2.3.2.13. Chromatographic conditions

The final chromatographic conditions established for the analysis of CP are summarized in Table

2.3.

Table 2.3. Chromatographic conditions for the analysis of CP

Column Nova-Pak® C18 150 mm x 3.9 mm i.d., 4 µm

Mobile phase Acetonitrile : water (50:50) Flow rate 1.0 ml/min Retention times 5.6 min (BV) and 8.2 min (CP) Column pressure 1300 psi Column temperature Ambient (22˚) Injection volume 15 µl Wavelength 240 nm Sensitivity 0.005 AUFS Integrator speed 0.5 mm/min Recorder input 10 mV full scale

2.4. METHOD VALIDATION

2.4.1. Overview

One of the most critical factors in developing pharmaceutical drug substances and products is

ensuring that the analytical test methods used to analyze fine chemicals and products generate

valid and meaningful data in terms of reliability, accuracy and precision, regardless of whether it

is intended for acceptance, release, stability or pharmacokinetic studies [81, 82]. Validation of

an analytical method is a process that provides documented evidence that an analytical test

method performs in an appropriate manner for the purposes for which it was intended [23, 83,

84].

The first step in the development of an HPLC method validation protocol is to determine the

objective of the intended method [82]. A method is considered a Level I or quantitative assay

method if it is intended to monitor blood levels in patients, for final product release, for the

determination of potency, or for the assessment of levels of impurity or contaminants in human

drug products [82]. If the method is to be used as a qualitative examination for identity, then it is

43

considered a Level II or qualitative assay method [81]. A variety of general validation protocols

have been recommended by organizations such as the FDA [84], United States Pharmacopeia

(USP) [23] and the International Conference on Harmonization (ICH) [85, 86].

The FDA [84], for example, requires a manufacturer of a pharmaceutical product to establish and

document accuracy, sensitivity, specificity and reproducibility of the analytical method [81]. The

USP [23] specifies typical performance characteristics such as accuracy, precision, specificity,

limit of quantitation (LOQ), limit of detection (LOD), linearity, range and sample stability that

should be considered in the validation of analytical methods intended for the analysis of active

ingredients alone or in finished pharmaceutical products, i.e., a Level I assay. Consequently, the

validation of an RP-HPLC analytical method for the analysis and characterization of CP in multi-

source topical products during formulation development and assessment was carried out as

outlined by the FDA [84] and USP [23] guidelines with specific reference to the ICH

recommendations [85, 86].

2.4.2. Linearity and Range

The linearity of an analytical method is an indication of the capability of a test method to

produce test results that are directly proportional to the amount of analyte in a sample within a

given concentration range [23, 83, 86]. Similarly, the working range of an analytical method

defines the inclusive upper and lower concentrations of an analyte in a sample for which it has

been demonstrated that the analytical procedure has a suitable level of precision, accuracy and

linearity [23, 83, 86].

The ICH guidelines [46] specify that a minimum of five (5) samples of increasing concentration

along with certain minimum specified ranges should be used to establish the linearity of a test

method [86]. The linearity of the analytical method was evaluated over the concentration range

0.10-18 µg/ml with the lower concentration representing the limit of quantitation and the upper

concentration indicating 120% of the test target concentration (15 µg/ml) in the samples to be

analyzed.

44

Calibration standards (0.10, 0.50, 1.0, 3.0, 6.0, 12 and 18 µg/ml) spiked with BV (IS) were

prepared as described in Section 2.3.2.9 and were injected (n = 5) onto the chromatographic

system described in Section 2.3.2.2 using the conditions described in Section 2.4.6. The peak

height ratios of CP response to BV response were calculated and used to construct a calibration

curve which was used to establish whether there was linearity of response of the analyte in

relation to concentration. Linearity was tested using least squares linear regression analysis of the

peak height ratios versus concentration data and the results of these studies are depicted in Figure

2.6.

y = 0.1061x - 0.0161R2 = 0.9994

0.0

0.5

1.0

1.5

2.0

2.5

0.0 5.0 10.0 15.0 20.0

Concentration (µg/ml)

Peak

hei

ght r

atio

Figure 2.6. Calibration curve constructed for CP following least squares linear regression analysis of peak height ratios of CP and IS versus concentration.

The acceptability of linearity data is determined by evaluation of the coefficient of determination

(R2) of the best fit linear regression line for the response vs. concentration plot as shown (Figure

2.6). An R2 value of greater than 0.999 is generally considered as evidence of an acceptable fit of

the data to the regression line [81]. The calibration curve in these studies was found to be linear

45

with an R2 value = 0.9994, a slope of 0.1061 and a y-intercept of -0.0161 yielding an equation

for the calibration line of y = 0.1061x – 0.0161.

2.4.3. Precision

The precision of an analytical method is a measure of the degree of scatter or closeness of

agreement among individual test results when the method is applied repeatedly to multiple

aliquots of the same homogeneous sample under the prescribed test conditions [23, 85].

Analytical method precision provides an indication of the variability of analytical results as a

function of the analyst, manipulation of samples and the day-to-day environment in which the

method is applied [87]. The precision of an analytical method is usually expressed as coefficient

of variation or percentage relative standard deviation (% RSD) of a series of measurements [23,

81, 85].

The FDA [84] recommends that the precision at each concentration level should not exceed 15%

RSD except for the LOQ, where it should not exceed 20% RSD for biological assays. It was

anticipated that the samples that would be collected in these studies would not be subject to high

degrees of variability and interference as would be expected for biological matrices and therefore

the acceptance criteria for precision studies was set at less than or equal to 5% RSD at each

concentration level, as opposed to the 15% RSD recommended by the FDA [84]. The ICH [85,

86] recommends that precision should be performed at three different levels i.e., repeatability,

intermediate precision and reproducibility.

2.4.3.1. Repeatability

Repeatability or intra-day precision of an analytical method is an indication of the performance

of an analytical procedure conducted within a laboratory over a short time interval using the

same analyst with similar equipment [23, 85]. The repeatability of this method was assessed by

calculating the % RSD of peak height ratios using five (5) replicates of the calibration standards

run on a single day. The results from repeatability studies (n = 5) are summarized in Table 2.4.

46

The data show in all cases that the % RSD values were less than 5%, indicating that the

analytical method for the analysis of CP was repeatable.

Table 2.4. Repeatability data for HPLC analysis of CP

Conc. (µg/ml) MPHR* (n = 5) SD % RSD 0.500 0.0498 0.00200 4.02 1.00 0.107 0.000100 0.900 3.00 0.332 0.00310 0.940 6.00 0.654 0.00470 0.710

12.0 1.18 0.00570 0.480 18.0 1.87 0.00570 0.300

* Mean peak height ratio of CP/internal standard

2.4.3.2. Intermediate precision

Intermediate precision or inter-day variability of an analytical method is an indication of the

variability in results obtained within a laboratory due to random events such as analysis on

different days, using different analysts and/or equipment [23, 81, 85]. The intermediate precision

of this method was assessed over a period of three consecutive days using five replicates of the

calibration standard concentrations prepared as previously described in Section 2.3.2.9.

Intermediate precision (n = 5) data expressed as coefficient of variation (% RSD) of the peak

height ratios of the calibration standards are shown in Table 2.5. The acceptance criteria for

intermediate precision in these studies was set at less than or equal to 5% RSD at each

concentration level and the data revealed that in all cases the % RSD values were less than 5%,

indicating that the analytical method would be precise when employed to analyze CP in semi-

solid dosage forms on different days.

47

Table 2.5. Intermediate precision data for HPLC analysis of CP

Day Conc. (µg/ml) MPHR* (n = 5) SD % RSD

0.500 0.0498 0.00200 4.02 1.00 0.107 0.00100 0.900 3.00 0.332 0.00310 0.940 6.00 0.654 0.00470 0.710

12.0 1.18 0.00570 0.480

1

18.0 1.87 0.00570 0.300 0.500 0.0474 0.00145 3.06 1.00 0.0974 0.000200 0.170 3.00 0.305 0.000600 0.200 6.00 0.561 0.00130 0.240

12.0 1.12 0.00320 0.270

2

18.0 1.81 0.00750 0.410 0.500 0.0569 0.00250 4.33 1.00 0.114 0.00320 2.70 3.00 0.292 0.000900 0.320 6.00 0.648 0.00140 0.220

12.0 1.24 0.00330 0.270

3

18.0 1.96 0.00400 0.200 * Mean peak height ratio of CP/internal standard

2.4.3.3. Reproducibility

The reproducibility of an analytical test method refers to the precision of a method following

application of that analytical procedure in different laboratories, and is often a part of inter-

laboratory crossover studies [23, 81]. Reproducibility studies should only be considered in the

case of standardized analytical procedures [86] and reproducibility studies are not normally

expected if intermediate precision is performed as part of method validation studies [81].

Consequently, reproducibility studies were not conducted, as the data for repeatability and

intermediate precision were considered sufficient to show that the analytical method was precise

and appropriate for the intended purpose of analyzing CP in semi-solid dosage forms.

2.4.4. Accuracy

The accuracy of an analytical method or procedure is an indication of the closeness of agreement

between a value measured or quantitated using the method and a value that is accepted to be

either a conventional true value or an accepted reference value for that sample [85]. In

48

combination precision and accuracy data determine the error of an analysis, and accuracy is

therefore an integral criterion in the validation of an analytical test method [83].

Accuracy may be determined in a number of different ways. For example, the accuracy of a

method can be determined by analyzing a sample of known concentration and comparing the

measured value to the true value or comparing test results from a new method with results from

an existing alternate well-characterized procedure or spiking a known amount of analyte in blank

matrices and calculating the percent recovery, [81]. Another approach used to assess the

accuracy of a method is to perform a two-sided t-test to determine if any significant differences

exist between the mean data generated by the test method and the nominal value, with a 95%

level of confidence [83].

The accuracy of the analytical method was determined by replicate analysis of samples

containing known amounts of CP. Three (3) samples representing low (1.5 µg/ml), medium (7.5

µg/ml) and high (15 µg/ml) concentrations prepared in mobile phase were injected in replicates

of five (5). Accuracy was reported as the percent recovery, % RSD and % Bias. Bias is the

difference between the mean value determined for an analyte of interest and the accepted true

value for that sample. Bias assesses the influence of the analyst on method performance and

accuracy measurements are designed to measure the effectiveness of sample preparation prior to

analysis [88].

The results from accuracy studies for this procedure are listed in Table 2.6. The acceptance

criteria for accuracy are that the mean percent recovered and the % RSD should be 100 ± 2%

[81] and less than 2% [89, 90], respectively, at each concentration level. A percent Bias of less

than 5% at each concentration level was considered as the test limit in our laboratory and based

on these criteria, the data in Table 2.6 reveal that this analytical method is accurate.

Table 2.6. Accuracy data for HPLC analysis of CP (n = 3)

Theortical conc. (µg/ml)

Actual conc. (µg/ml)

SD % RSD

% Bias

% Recovery

1.50 1.48 0.005000 0.360 -1.61 98.67 7.50 7.47 0.0160 0.210 -0.360 99.60

15.0 15.1 0.00300 0.0200 +0.460 100.7

49

2.4.5. Limit of quantitation (LOQ) and limit of detection (LOD)

The limit of quantitation (LOQ), sometimes referred to as the lower limit of quantitation

(LLOQ), is the lowest concentration of an analyte in a sample that can be determined with

acceptable precision and accuracy under the stated operational conditions of an analytical

method [83, 91]. Similarly, the limit of detection (LOD) is the lowest concentration of analyte in

the sample that can be detected but not necessarily quantitated under the stated experimental

conditions of the analytical method [23, 83, 91]. The LOQ is normally taken as the lowest

concentration point in the calibration curve [83] whereas the LOD is a limit test that merely

specifies whether or not the amount of an analyte is above or below a certain level [23, 91].

Four methods are commonly used to determine the LOQ and LOD of an HPLC method [83, 91].

The ultimate method of choice is usually left to the discretion of the analyst or the standard

operating procedures of a specific laboratory, since all four techniques essentially yield

equivalent results and are suitable for satisfying the USP [23] and ICH [85, 86] requirements for

the determination of the LOQ and LOD of an analytical method [91]. Thus, the LOQ in these

studies was determined by evaluating the lowest concentration of analyte that resulted in a

precision of less than 5% RSD [91], in other words the LOQ is the lowest concentration for

which the % RSD of multiple injections of a sample (n = 6) was less than 5%. By convention the

LOD value is taken as 30% of the LOQ value [91].

Five different concentrations of CP were evaluated as potential LOQ values and the data

generated in these studies are depicted in Table 2.7. Based on these results the LOQ for this

analytical test method was found to be 0.10 µg/ml, with a % RSD value of 1.82% and by

convention, the LOD value was taken as 0.03 µg/ml, which when injected into the HPLC,

resulted in a detectable but not quantifiable peak.

50

Table 2.7. LOQ data for HPLC analysis of CP

Conc. (µg/ml) MPHR* (n = 6) SD % RSD 0.25 0.0312 0.000200 0.670 0.20 0.0261 0.000300 1.11 0.15 0.0210 0.000400 1.70 0.10 0.0154 0.000300 1.82 0.050 0.0133 0.00140 10.8


2.4.6. Specificity and selectivity

2.4.6.1. Overview

Specificity and selectivity are relative terms, often used interchangeably to describe the ability of

an analytical test method to measure accurately the analyte of interest in the presence of other

components that may be present in a sample matrix [88, 92]. However, an analytical method is

specific if it produces a response for only a single analyte, whereas selectivity describes a

procedure that provides a response for the target compound that is distinguishable from all other

responses that may be generated from a sample matrix [92, 93].

As most chromatographic procedures produce responses for other substances and not only for the

analyte of interest, the term selectivity is more appropriate than specificity in this context [93].

Consequently, the term selectivity is applied for these studies. Selectivity was evaluated by

analyzing a sample of a commercially available 0.05% w/w CP cream product (Dermovate®,

Glaxo Wellcome SA (Pty) Ltd, Midrand, RSA) and an extemporaneous placebo cream

formulation, using a simple liquid-liquid extraction procedure with hexane and acetonitrile,

which is described in Section 2.4.6.2.

2.4.6.2. Sample preparation

A schematic representation of the sample preparation procedure is shown in Figure 2.7.

Approximately 600 mg of a cream sample, equivalent to 0.30 mg of CP, was weighed and

transferred into a 100 ml Duran® Schott round neck Erlenmeyer flask (Schott Duran GmbH,

51

Hattenbergstrasse, Germany). A standard stock solution for BV was prepared in ACN as

previously described in Section 2.3.2.9 and used to prepare 7.5 µg/ml BV in ACN. A 20 ml

aliquot of the BV solution was added to a stoppered flask containing the cream, followed by the

addition of 30 ml of n-hexane (Associated Chemical Enterprises (Pty) Ltd, Johannesburg,

Gauteng, RSA).

The mixture was shaken vigorously to disperse the semi-solid sample and sonicated for 10

minutes to aid the dissolution of CP in ACN using a Model 8845-30 ultrasonic bath (Cole-

Parmer Instrument Comp., Chicago, IL, USA). The mixture was then transferred to a Model-NS

19/26, 1.5 x 12.5 mm stopcock separating funnel (VIT-LAB GmbH, Seeheim-Jugenheim,

Germany) and was shaken vigorously while venting the funnel to avoid a potential build up of

pressure within the funnel.

Following the separation of the immiscible solvents, the lower distinct ACN layer was

withdrawn via the stopcock of the separating funnel and was collected in a clean Duran® Schott

round neck Erlenmeyer flask. The sample was then filtered through a 0.22 µm Millipore filter

(Millipore Co., Bedford, MA, USA) and an aliquot of the filtered sample was injected onto the

chromatographic system in replicates of six (6). A calibration curve was constructed on the same

day that the samples were prepared and analysed, and the concentration of CP in the sample

filtrate was interpolated from the calibration curve.

52

Figure 2.7. Schematic representation of the sample preparation procedure

600 mg Cream

20 ml 7.5 µg/ml BV solution made in

acetonitrile (ACN)

30 ml n-hexane

Shake mixture vigorously

Sonicate mixture for 10 minutes

Separating funnel

Shake mixture vigorously

Collect ACN layer

Filter through 0.22 µm

Millipore filter

Filtrate HPLC

53

2.4.6.3. Extraction efficiency

A commercially available 0.05% w/w CP cream product (Dermovate®, Glaxo Wellcome SA

(Pty) Ltd, Midrand, SA) was used to evaluate the extraction efficiency or percentage recovery of

the sample preparation procedure described in Section 2.4.6.2. The extraction efficiency was

determined using 600 mg samples of the cream (n = 4). Based on the labelled amount of CP in

the semi-solid dosage form (0.05% w/w CP) and assuming that the entire CP in the cream

sample dissolved in the 20 ml ACN added, a theoretical concentration of 15 µg/ml was expected.

The theoretical concentration was used as a reference sample to determine the actual

concentration of CP in the sample filtrate for the calculation of the percentage recovery values.

The extraction efficiency and precision data generated from these studies are tabulated in Table

2.8. The United States Pharmacopoeia (USP) [23] specifies that CP cream formulations should

contain not less than 90.0% and not more than 115.0% of the labelled amount of CP. The

extraction efficiency data shown in Table 2.8 fall within this range with excellent precision and

therefore, the cream formulation tested in these studies complied with the USP standards.

Section 2.4.6.4 describes the validation of the extraction procedure.

Table 2.8. Extraction efficiency data following the extraction of CP from Dermovate® cream (n = 6)

Sample weight (mg)

Expected theoretical

conc. (µg/ml)


SD % RSD % Recovery

608.9 15.22 14.85 0.04700 0.3200 97.53 600.6 15.01 14.23 0.1314 0.9200 94.77 601.8 15.05 14.43 0.08290 0.5700 95.94 594.1 14.85 14.35 0.1387 0.9700 96.60

2.4.6.4. Validation of the extraction procedure

In order to determine the authenticity of the sample preparation procedure described in Section

2.4.6.2 and to ensure that the data generated in the studies described in Section 2.4.6.3 and

summarized in Table 2.8 were valid, the method was applied to the extraction of CP from a

0.05% w/w solution. The 600 mg sample of CP cream was replaced by 600 µl of a CP solution.

54

Samples (n = 3) of the CP solution were extracted and analysed and the data generated in these

studies are summarized in Table 2.9.

Table 2.9. Extraction efficiency following extraction of 600 µl of a CP solution (0.05% w/w) (n = 3)

Sample volume

(µl)

Expected theoretical

conc. (µg/ml)


SD

% RSD

% Recovery

600.0 15.00 14.98 0.04800 0.3200 99.84 600.0 15.00 14.23 0.1314 0.9200 94.77 600.0 15.00 15.86 0.03352 0.1200 105.8

The data show excellent percentage recovery and precision values that are similar to those

obtained when CP cream was used (Section 2.4.6.3) and are indicative of the suitability of the

method used to prepare samples of the cream, for analysis. The extraction process was therefore

used in subsequent studies to determine and confirm CP content in extemporaneously prepared

0.05% w/w creams during product development and assessment studies.

2.4.6.5. Selectivity studies

The selectivity of the analytical method was evaluated by analyzing a sample of an

extemporaneous placebo cream formulation, extracted with:

a) ACN without BV, and

b) ACN spiked with the BV.

The selectivity was further evaluated by analyzing a sample of a Dermovate® cream extracted

with:

c) ACN without BV, and

d) ACN with BV.

Typical chromatograms obtained in these studies are shown in Figure 2.8. Inspection of the

resultant chromatograms reveal that the CP and BV peaks are free from interference from other

formulation excipients, including chlorocresol (CH), which was the preservative in the

commercially available CP cream. The method was therefore considered selective for the

purposes of analyzing CP in semi-solid formulations.

55

121086420


CH

inje

ctio

n

121086420


BV

CH

inje

ctio

n

121086420


CP

CH

inje

ctio

n

121086420


CP

BV

CH

inje

ctio

n

solv

entfr

ont

solv

entfr

ont

solv

entfr

ont

solv

entfr

ont

A B

C D

Figure 2.8. Typical chromatograms obtained following the analysis of a sample of placebo cream without BV (A) and with BV (B) and Dermovate® cream without BV (C) and with BV (D)

56

2.4.7. Sample stability

2.4.7.1. Overview

Another important consideration during method validation is the demonstration of stability of an

analyte in a sample matrix and solvents used during the sample work-up, under the conditions to

which the study samples will be subjected throughout the analytical procedure [93]. It has been

reported that a compound may be considered stable under certain conditions for a certain period

of time when there is no significant difference between the detector response for an analyte

following analysis of solutions that have been stored, compared to the response generated from

freshly prepared solutions of the analyte in similar matrices [83].

There are numerous areas that need to be identified and stability evaluations undertaken during

sample analysis [83, 93]. In these studies, the stability of CP in the standard stock solutions

(Section 2.3.2.9) used to prepare the calibration standards (Section 2.3.2.9) and the stability of

CP in process samples used during in vitro release studies (Chapter 3) were evaluated. It has

been recommended that the stability of standard stock solutions of an analyte used to prepare

calibration curves be evaluated over the maximum time period for which the solutions will be

stored prior to use [83, 93, 94]. Accordingly, the stability of the stock solutions must be

determined under the same conditions of light or dark, at the same temperature(s), and in the

same solvent(s) and container(s) as are used during analysis [83, 93].

Buick et al., [94] suggested that in the absence of stability data, standard stock solutions must be

freshly prepared on a daily basis during sample analysis, since failure to do so may lead to the

preparation of calibration standards that do not represent the true concentrations of the analyte,

which may then result in the determination of incorrect concentrations in unknown samples.

Similarly, in-process sample stability studies are required in order to determine the stability of a

compound in a sample matrix under specific storage conditions over the time needed to store and

analyze the samples [95, 96]. The objective of conducting these stability studies was to

determine if any degradation of CP occurs during the entire period of sample collection,

processing, storing, preparation and analysis [83]. Consequently, the data obtained from in-

57

process sample stability studies may be used to provide guidelines concerning the maximum

times and conditions of sample storage prior to analysis and therefore to ensure that the integrity

of samples is maintained up to the point of analysis [93, 94, 96].

2.4.7.2. Stability data analysis

Stability data generated in these studies were analysed using a statistical test method developed

by Timm et al., [97]. The procedure was initially developed for the evaluation of stability of

drugs in biological fluids and takes into account the quality of the experimental procedure

including the precision of the method and the number of replicates used during analysis of the

stored samples [97]. The statistical interpretation of the stability data as proposed by Timm et al.,

[97] is based on the construction of a 90% confidence interval (C.I.) for the true percentage

change (∆) in concentration or detector response generated from stored and freshly prepared

samples.

The lower limit (L.L.) and the upper limit (U.L.) of the C.I. are calculated using the measured

percentage response difference (D) between stored and freshly prepared samples for replicate

analyses and the true percentage change in response lies within that limit with 90% certainty

[97]. According to Timm et al., [97] a change of response during storage may be considered

statistically relevant if the values for both the L.L. and U.L. of the C.I. are either < -10 or > 10.

Figure 2.9 depicts the possible outcomes that could be generated when stability data are analyzed

using the method described by Timm et al., [97].

58

a

b

c

d

e

f

LL UL

-20 -10 0 10

% Change from initial concentration

20

The bars above the axis depict the ranges of the 90% CI for the % ∆ between stored and freshly prepared samples.

a) change of response, not significant and not relevant b) decrease of response, significant, but not relevant c) decrease of response, significant and possibly relevant d) decrease of response, significant and relevant e) decrease of response, not significant, but possibly relevant f) increase of response, significant

Figure 2.9. Interpretation of stability data, as described by Timm et al, [97]

2.4.7.3. Stability of stock solutions

The stability of CP stock solutions prepared in ACN was evaluated following storage for two

weeks at 4˚C. CP stock solutions were prepared as previously described in Section 2.3.2.9. 75 µl

and 1500 µl aliquots of CP stock solutions were measured using single channel model electronic

pipettes covering the volume range of 50-1200 µl and 500-5000 µl respectively (ePET Electronic

Pipettes Biohit, Helsinki, Finland), and transferred to a 10 ml A-grade volumetric flask and made

up to volume with ACN, yielding solutions of 0.75 µg/ml and 15 µg/ml, representing the lower

and upper concentrations, respectively.

59

A maximum of five (5) replicate samples at each concentration studied were prepared for

analysis on days 1, 2, 3, 7 and 14 after storage at 4˚C, resulting in a total of twenty-five (25)

separate samples being analyzed. On each day of analysis fresh samples at each concentration

were prepared from a freshly made CP stock solution (Section 2.3.2.7) and analysed together

with the five (5) stored samples. Both freshly made and stored samples were spiked with 60 µl of

a freshly prepared 0.5 mg/ml BV solution (measured using a 50-1200 electronic pipette) prior to

analysis, and for each sample analyzed, the peak height was measured and the ratio of CP to BV

was used as the response.

The results obtained from stability studies of CP stock solutions at both the lower and upper

concentrations stored at 4˚C for 1, 2, 3, 7 and 14 days are depicted in Figure 2.10. These data

reveal that at the lower (0.75 µg/ml) and upper (15 µg/ml) concentrations, stored at 4˚C for 14

days the change of response for CP was neither significant nor relevant. The wide confidence

interval calculated for the 0.75 µg/ml sample on days 2 and 3 may be indicative of the poor

precision of the assay procedure at these low levels [97]. Nevertheless, these studies reveal that

CP was stable in acetonitrile at 4˚C following storage for two weeks. Consequently, stock

solutions of CP were prepared in acetonitrile, stored at 4˚C and used within a period of two

weeks, after which fresh solutions were made.

60

-20 -10 0 10


20

3 Days at 4 Co

7 Days at 4 Co

14 Days at 4 Co

KEY

0.75 g/ml� 15 g/ml�

2 Days at 4 Co

1 Day at 4 Co

Figure 2.10. Stability of CP in ACN at two different concentrations, stored at + 4ºC for 1, 2, 3, 7, and 14 days

61

2.4.7.4. In-process sample stability

In-process sample stability studies were also conducted by evaluating CP stability in the receptor

medium to be used for in vitro release studies. A receptor medium of 50% v/v propylene

glycol/water (Chapter 3, Section 3.2.6.2.3.3) was to be used during the in vitro release studies at

room temperature (22˚C). Stock solutions of CP were prepared as previously described in

Section 2.3.2.9 but in this case propylene glycol rather than ACN was used as the solvent of

choice. A 75 µl and 1500 µl aliquot of the CP stock solution measured as described in Section

2.5.7.3 was diluted and made up to 10 ml using the receptor phase, yielding concentrations of

0.75 µg/ml and 15 µg/ml concentrations as previously described in Section 2.5.7.3.

A maximum of five (5) replicate samples were prepared at both the lower and upper

concentration and stored for 1, 2 and 3 days, resulting in a total of fifteen (15) separate samples

for each concentration being studied. The samples were stored at room temperature (22˚C) and

on each day of analysis stored and freshly prepared samples (n = 5) spiked with 60 µl of a freshly

prepared 0.5 mg/ml internal standard (BV) were analysed. The percentage response differences

between these samples were calculated and used to construct a 90% CI as described by Timm et

al., [97] and the data generated from these studies are depicted in Figure 2.11.

The results indicate that the change of response of CP at both the lower (0.75 µg/ml) and upper

(15 µg/ml) concentrations prepared in 50% v/v propylene glycol/water and stored at room

temperature (22˚C) for three (3) consecutive days was not significant and not relevant. CP was

therefore considered to be stable when prepared in 50% v/v propylene glycol/water and stored at

room temperature (22˚C) for three days. Consequently, CP samples from in vitro release studies

were stored at 22˚C after removal of each sample and analyzed immediately at the conclusion of

the in vitro release studies, i.e. within three days of the commencement of the study.

62

-20 -10 0 10


20

1 Day at 22 Co

2 Days at 22 Co

3 Days at 22 Co

KEY

0.75 g/ml� 15 g/ml�

Figure 2.11. Stability of CP in 50 %v/v propylene glycol/water stored at + 22ºC for 1, 2 and 3 days

2.5. METHOD REVALIDATION

2.5.1. Overview

According to the FDA guidance document on Analytical procedures and methods validation:

Chemistry, manufacturing and controls documentation [98], it is vital that an analytical method

is revalidated when changes are made to an original procedure or to the original operating

conditions of a validated method. Such revalidation studies are intended to ensure that the

method maintains the appropriate performance characteristics as specified prior to the

implementation of the changes to the method [95, 98].

Although the extent of the revalidation procedure to be undertaken depends on the nature of the

change to a method, it is not necessary to conduct extensive revalidation studies when an

63

analytical method has been previously fully validated according to an international protocol [99].

Therefore, the decision concerning which parameters require revalidation is based on logical

consideration of those specific validation attributes that are more than likely to be affected by the

change in question [95]. However, as a minimum requirement, linearity, precision and accuracy

studies should be undertaken in most cases [99].

The revalidation of this analytical method was necessary since a change was made to the solvent

delivery module of the system as outlined in Section 2.3.2.2. The definitions and acceptance

criteria of the relevant performance characteristics discussed below have been previously

described in Section 2.4.2.

2.5.2. Linearity

A calibration curve was constructed by plotting the peak height ratios of CP to BV versus CP

concentration within the concentration range 0.10-18 µg/ml and performing least squares linear

regression of the constructed calibration curve. The data revealed that the calibration curve was

linear with an R2 value = 0.9999, a slope of 0.1093 and a y-intercept of 0.0026 yielding a curve

of y = 0.1357x + 0.0015 and this is depicted in Figure 2.12.

64

y = 0.1093x - 0.0026R2 = 0.9999

0

0.5

1

1.5

2

2.5

0 5 10 15 20

Concentration (µg/ml)

Peak

Hei

ght r

atio

Figure 2.12. Calibration curve for CP following revalidation of the method

2.5.3. Precision

2.5.3.1. Repeatability

Repeatability was determined by calculating the coefficient of variation of the peak height ratios

of CP to BV of the calibration standards (n =5). The data summarized in Table 2.10 revealed that

in all cases, the relative standard deviations were less than 5%, indicating that the analytical

method for the analysis of CP was precise in terms of the repeatability criterion.

65

Table 2.10. Repeatability data for the revalidation of the HPLC method for CP

Conc. (µg/ml) MPHR* (n = 5) SD % RSD 0.250 0.0260 0.000400 1.47 1.00 0.127 0.000600 0.460 6.00 0.883 0.00180 0.200

12.0 1.56 0.00230 0.150 18.0 2.47 0.00580 0.230


2.5.3.2. Intermediate precision

Intermediate precision was assessed over a period of three consecutive days and was expressed

as coefficient of variation (% RSD) of the peak height ratios of the calibration standards (n = 5).

The data summarized in Table 2.11 revealed that all % RSD values were less than 5%, indicating

that the analytical method conforms to the requirements for intermediate precision.

Table 2.11. Intermediate precision data for the revalidation of the HPLC method for CP

Day Conc. (µg/ml) MPHR* (n = 5) SD % RSD

0.250 0.0260 0.000400 1.47 1.00 0.127 0.000600 0.460 6.00 0.883 0.00180 0.200

12.0 1.57 0.00230 0.150

1

18.0 2.47 0.00580 0.230 0.250 0.0281 0.000400 1.59 1.00 0.128 0.000400 0.330 6.00 0.720 0.000300 0.050

12.0 1.64 0.00170 0.100

2

18.0 2.52 0.00120 0.0500 0.25 0.0326 0.000700 2.08 1.00 0.133 0.000300 0.260 6.00 0.712 0.000200 0.0300

12.0 0.542 0.00110 0.0700

3

18.0 2.59 0.00350 0.140 * Mean peak height ratio of CP/internal standard

2.5.3.3. Reproducibility

Reproducibility studies were not conducted for the same reasons as discussed in Section 2.4.3.3.

66

2.5.4. Accuracy

Accuracy was assessed at three concentrations viz., 1.5 µg/ml (low), 7.5 µg/ml (medium) and 15

µg/ml (high), (n = 5) and was reported as % recovery, % RSD and % Bias. The data summarized

in Table 2.12 show that the analytical method can be considered accurate for the analysis of CP

in these studies.

Table 2.12. Accuracy data for the revalidation of the HPLC method for CP (n = 3)

Theortical conc. (µg/ml)


SD % RSD

% Bias

% Recovery

1.50 1.46 0.00200 0.120 -2.39 97.33 7.50 7.47 0.0140 0.230 -0.360 99.60

15.0 14.6 0.0270 0.190 +2.84 97.20

2.6. APPLICATION OF THE ANALYTICAL METHOD

Following development and validation studies, the RP-HPLC method was applied to the

quantitative determination of CP in samples obtained following in vitro release testing of CP

proprietary cream product (Dermovate®, Glaxo Wellcome SA (Pty) Ltd, Midrand, RSA). A

typical in vitro release profile (n = 6) generated during these studies is depicted in Figure 2.13.

67

0

5

10

15

20

25

30

0 20 40 60 80

Time (hr)

% R

elea

sed

Figure 2.13. In vitro release profile of CP

2.7. CONCLUSIONS

A suitable RP-HPLC analytical method with UV detection at 240 nm has been developed,

optimized and validated for use in the in vitro analysis of CP in topical formulations. Method

development studies included the selection of a suitable detection wavelength, analytical column,

and internal standard (I.S.), as well as the selection of mobile phase components.

The ultraviolet (UV) absorption spectrum of CP revealed that the wavelength of maximum

absorption (λmax) of CP is 240 nm. Consequently, the UV detector in these studies was set at 240

nm. The selection of the analytical column was based on the physicochemical properties of CP,

the type and size of column packing materials, column dimensions and the column efficiency as

determined by the number of theoretical plates (N). Based on these features a Nova-Pak® C18 4

µm (150 mm x 3.9 i.d.) cartridge column was selected as a suitable column for the analysis of

CP.

68

Betamethasone 17-valerate (BV) was selected as the most suitable I.S. from amongst the various

compounds that were tested. Analysis using RP-HPLC is best achieved with non-polar stationary

phases such as n-octadecylsilane (ODS or C18) and with mobile phases consisting of a polar

solvent i.e. water or a mixture of water and water-miscible organic modifier such as methanol

(MeOH) and/or acetonitrile (ACN). A binary mixture of water and an organic modifier was

considered the most suitable mobile phase composition and ACN was chosen over MeOH as the

most appropriate organic modifier.

The method was optimized by manipulation of ACN concentration in the mobile phase and

mobile phase flow rates of the optimal mobile phase composition selected. The effects of mobile

phase composition and flow rate on retention time and resolution of CP and BV were assessed.

The separation of CP and BV was best achieved using a mobile phase composition consisting of

a binary mixture of ACN-water (50:50) and a mobile phase flow rate of 1 ml/min. These

chromatographic conditions resulted in retention times of 5.6 min and 8.2 min for BV and CP

respectively, and the total run time for sample analysis was within 10 min.

The RP-HPLC analytical method was validated in terms of the guidelines recommended by

various organizations such as the United States Pharmacopeia (USP) [23], the United States

Food and Drug Administration (FDA) [84] and the International Conference on Harmonization

(ICH) [85, 86]. Based on these guidelines, the analytical method was found to be linear, precise,

accurate, selective and sensitive and CP was found to be stable in acetonitrile and 50% w/w

propylene glycol-water binary mixture following storage at 4˚C and room temperature (22˚C),

respectively, for a maximum of 14 days and 3 days, respectively.

Following changes to the modular HPLC system used to develop, optimize and validate the RP-

HPLC analytical method described herein, a mini-revalidation of the analytical method was

carried out so as to ensure that the method maintained its performance characteristics as reported

prior to the implementation of the changes. The data from revalidation studies revealed that the

analytical method was linear, accurate, and precise for the in vitro analysis of CP in semi-solid

dosage forms.

69

A RP-HPLC has been developed, optimized and validated RP-HPLC for the quantitation of CP

in cream formulations and can be applied to the assessment of in vitro performance of CP in

pharmaceutical semi-solid dosage forms

70

CHAPTER THREE

DEVELOPMENT AND VALIDATION OF AN IN VITRO TEST METHOD FOR THE

ASSESSMENT OF CLOBETASOL 17-PROPIONATE RELEASE FROM TOPICAL

CREAM FORMULATIONS

3.1. INTRODUCTION

In vitro dissolution testing of solid oral dosage forms is a well-established technique used to

assess drug release from tablets and capsules for the purposes of assessing quality [100, 101].

Dissolution testing is considered the single most valuable in vitro test method that can be used to

guide formulation development, assess product quality and ensure batch-to-batch uniformity

[102]. In addition, in vitro dissolution testing may be used as a pre-formulation tool and it has

been suggested that dissolution testing is a useful surrogate measure for the prediction of

bioavailability for extended-release oral dosage forms using pre-established in vitro-in vivo

correlations (IVIVC) [101, 103].

Official dissolution test methods have been developed and reported for in vitro dissolution

studies of solid oral dosage forms and for the investigation of in vitro release characteristics of

active pharmaceutical ingredients (API) from transdermal patches [22, 23]. However, there are

currently no official guidelines or requirements for the performance evaluation of drug release

from semi-solid dosage forms [104, 105]. Nevertheless, a variety of physical and chemical

quality control tests have been traditionally used to provide reasonable evidence of consistent

product performance [49]. These tests include the determination of solubility, particle size and

size distribution and crystalline form of an API and evaluation of the intrinsic viscosity and

homogeneity of a final product [49, 104-108].

These tests, however, provide little information about the drug release characteristics from the

product or the effects of processing and manufacturing variables on the performance of a

finished product [106]. As an API must first be released from a formulation and then permeate

through the stratum corneum in order to exert a therapeutic effect [109], it may be more

71

appropriate to use in vitro drug release characterization in conjunction with traditional quality

control tests to determine the quality and consistency of topical formulations that are

manufactured during product development studies [109].

The FDA issued a Guidance document [49] in 1997, in which issues relating to scale-up and

post-approval formulation changes of semi-solid dosage forms were addressed [49]. This

guidance document outlines the requirements that the pharmaceutical industry must meet to

maintain certification of previously approved semi-solid dosage forms, following post-approval

formulation changes [49]. A key element in the guidance document is the requirement that an in

vitro release test method be used to determine if the diffusional rate of release of a drug from a

formulation is the same following changes to that formulation as it was prior to the

implementation of the changes being made [49].

Despite the issuance of this Guidance document [49] by the FDA, an official in vitro release test

is still not mandatory and is not applied industry-wide as a quality control test when compared to

the utility and requirements for in vitro dissolution testing of solid and liquid oral dosage forms

[107]. This is more than likely due to the fact that a universally acceptable in vitro release test

protocol that can be applied to all semi-solid dosage forms has yet to be established [108, 110].

With the exception of the recommendations of the FDA, there are currently no apparatus,

procedures or requirements that have been described in any pharmacopoeias for in vitro release

test of topical semi-solid preparations [103, 107, 110, 111].

Another point relevant to in vitro release testing is that it cannot be considered as a surrogate test

for establishing bioequivalence of a generic product relative to an innovator semi-solid

formulation [49, 107, 112]. This is more than likely due to the fact that at this time there is no

convincing evidence for IVIVC of release tests for semi-solid dosage forms as that for in vitro

dissolution of tablets and capsules [49]. Moreover, measurement conditions during in vitro

release testing do not usually mimic physiological reality since in vitro release test equipment

does not include a membrane resembling the stratum corneum and the barrier functions of the

skin, which are essential determinants of the skin penetration characteristics of an API of interest

in semi-solid dosage form bases [113].

72

The type of membrane usually chosen for in vitro release testing should have the least possible

diffusional resistance, whereas, clinically dosage forms are applied to the stratum corneum, a

high resistance membrane that, when intact invariably controls the delivery rate of API from

semi-solid dosage forms [112]. Furthermore it should be noted that in vitro release testing is

performed in a manner in which the dosage form being tested is applied so that an infinitely thick

layer is present, which is in stark contrast to the thin films usually applied during conventional

clinical use of these products [114].

Nevertheless, many manufacturers of topical drug products have devoted significant resources to

developing and validating in vitro release tests during the product development process [108].

Such tests can be used to detect the effects of changes in a formulation on the release rate of API

from a dosage form in which the API is suspended and/or dissolved and therefore can be used to

ensure that the manufacture of semi-solid products is consistent and fulfils a similar role as

dissolution testing does for tablets and capsules [115].

The objective of these studies was therefore to develop and validate a reliable, reproducible and

discriminatory in vitro release test method. The method would be applied in formulation

development studies in order to assess product quality and ensure batch-to-batch consistency of

topical formulations manufactured to contain 0.05% w/w clobetasol 17-propionate (CP).

73

3.2. METHOD DEVELOPMENT

3.2.1. Overview

Various protocols that may be used in the development of an in vitro release test method for a

semi-solid drug product have been outlined and reported in the literature [108, 113, 115].

Essentially, method development studies are designed to:

a) facilitate the selection of a diffusion cell system,

b) determine an appropriate receptor medium with adequate sink characteristics,

c) evaluate synthetic support membranes with minimal reactivity with the drug and

formulation components, and

d) determine appropriate sampling times [108, 113, 115].

In addition, conditions such as finite or infinite dosing and sample occlusion or non-occlusion

that may have considerable effects on the in vitro release characteristics of a drug should be

investigated and optimized during method development studies [115]. During in vitro release

method development studies, the aforementioned criteria were evaluated to produce a test system

for the assessment of CP release from topical semi-solid formulations.

3.2.2. Diffusion cell test system

3.2.2.1. Overview

In vitro dissolution testing in the pharmaceutical industry is a widely used tool in formulation

development and quality control testing [111]. Consequently as dissolution testing has evolved

different apparatus have become official and have been included in the United States

Pharmacopoeia (USP) [23] and the British Pharmacopoeia (BP) [22] for such purposes. The USP

[23] for example makes provision for the dissolution testing of tablets and capsules, specifying

and describing official apparatus as USP Apparatus I (basket), USP Apparatus II (paddle), USP

Apparatus III (reciprocating cylinder or Bio-Dis®) and USP Apparatus IV (flow-through). In

74

addition, the USP [23] specifies and describes methods and equipment for transdermal delivery

patch systems such as USP Apparatus V (paddle over disk), USP Apparatus VI (cylinder) and

USP Apparatus VII (reciprocating holder).

Although the FDA Guidance document [49] outlines the general methodology and description of

diffusion systems, no single device has been universally accepted for measuring in vitro drug

release rates from semi-solid dosage forms. Due to the variety of formulation type, sites of

application and release rates required from semi-solid dosage forms, a single test method would

not be appropriate for the development, biopharmaceutical characterization and quality control of

such formulations, and hence the inclusion of a single apparatus in pharmacopoeias may not be

possible [111].

Numerous apparatus of different design have been reported in the literature for studying API

diffusion from semi-solid dosage forms [23, 49, 102-107, 109-131]. The general design of the

test systems is such that they consist of a donor cell in which a semi-solid material is placed and

a receptor compartment in which a chosen receptor medium is contained and from which

samples are withdrawn for analysis at appropriate intervals. The system may or may not include

a membrane that separates the two compartments and would also include a mechanism by which

the receptor fluid is stirred and a means to control the temperature of the receptor medium [113].

Several reviews have alluded to the many significant considerations that must be taken into

account when designing such test devices [113, 115, 119, 124]. An in vitro diffusion cell must be

made from an inert, non-reactive material such as, for example, glass, stainless steel or Teflon®

[122]. All components of a cell, including flow-through lines and the collection chambers, must

have their inert nature established by experimental means and it should be shown that there is no

loss of drug due to volatility during the permeation test procedure [122]. If volatility of an API or

receptor and/or product being tested is an issue, a quantitative account of the volatility must be

undertaken and reported [113, 122].

A receptor medium must provide effective sink conditions for the appropriate assessment of the

rate of release of a permeant, penetrant or API [115]. The cell in which the medium is placed

75

should ideally contain a minimum volume of fluid to facilitate analysis, since the more

concentrated a drug in the collection medium is, the easier the analytical process [122]. The

design of a cell must allow for efficient mixing of the receptor fluid at a controlled temperature

[113, 122] and the system should be versatile and accommodate a diverse range of membranes

that are routinely used in diffusion studies [124]. The current trend is towards the development

and use of systems with automated operations with better instrument control and minimal

instrument-related variability [128].

3.2.2.2. Franz diffusion cell

Studies using a diverse range of in vitro release systems have been reported [23, 49, 102-107,

109-131]. Of those reported the vertical diffusion cell, commonly known as the Franz and/or

modified Franz cell has shown the most potential for use as a standardized system that may be

adapted for use as a compendial test method [111, 113]. Initially designed and described by

Franz in 1975 [132], the vertical Franz diffusion cell has been widely used to study in vitro drug

release from semi-solid dosage forms [102, 107, 114-118, 121, 123, 126, 127, 129, 133-135].

The original Franz cell has a bichamber arrangement that includes a dumb-bell-shaped receptor

compartment [133] and a schematic of this apparatus is shown in Figure 3.1. An unstoppered

sampling port is connected to the upper section of the receptor chamber and only the central,

cylindrical portion of the receptor compartment is jacketed. A stirrer bar is placed in the lower,

ellipsoid bulb chamber to provide a means of agitation of the receptor medium.

76

68.5mm

27mm

28mm

13.5mm

12.5mm

17mm

30mm

RECEPTOR COMPARTMENT(cell body)

WATER JACKET

STIRRING BAR

MEMBRANE

SAMPLING PORT

SEMI-SOLID SYSTEM

DONOR COMPARTMENT(cell cap)

20mm

AIR

Figure 3.1. Schematic representation of an original Franz diffusion cell apparatus (redrawn from 127)

3.2.2.3. Modified Franz diffusion cell

Keshary and Chien [127, 136] identified and reported deficiencies associated with use of the

original Franz diffusion cell. They reported that the design of the Franz cell did not provide for

adequate solution hydrodynamics, mixing efficiency and temperature control that are necessary

for quantitative permeation studies. Due to these shortcomings, Keshary and Chien [127, 136]

proposed several modifications to the original design of the Franz cell. A schematic

representation of a modified Franz diffusion cell is illustrated in Figure 3.2.

The diffusion cell has a cylindrical receptor compartment that is shorter than the receptor

compartment of the original Franz cell (Figure 3.1) and is completely enclosed by a water jacket.

A star-head magnetic stirrer as opposed to a stirrer bar is used to agitate the receptor fluid. As a

result of these modifications, the equilibrium temperature of the receptor fluid is better

77

maintained, while solution-mixing efficiency is improved and the thickness of boundary

diffusion layer is reduced [127].

SAMPLING PORT

38mm

50mm

20mm

30mm

20mm

AIR

STAR-HEAD MAGNET

DONOR COMPARTMENT(cell cap)

SEMI-SOLID SYSTEM

MEMBRANE

RECEPTOR COMPARTMENT(cell body)

WATER JACKET

Figure 3.2. Schematic representation of a modified Kenshary-Chien Franz glass diffusion cell (redrawn from 127)

3.2.2.4. Selection of diffusion cell test system

A modified Franz glass diffusion cell system (Crown Glass Company Inc., Branchburg, NJ,

USA) was selected for use in these studies. Each jacketed cell has an opening of 15 mm with an

associated area available for diffusion of 1.767 cm2 and a receptor fluid volume of 12.5 ml. The

cells were positioned in a multiple-cell drive unit (Figure 3.3), which permitted efficient stirring

with a 2x2 mm star head magnetic stirrer bar (Merck Chemicals (Pty) Ltd, Darmstadt, Germany)

agitating the receptor medium at a controlled rate. The jacketed portions of the multiple cells

were connected in series to a temperature-controlled circulating water bath (Grant Instruments

Ltd, Cambridge, England) with Tygon® tubing (Eagle Electric, Cape Town, Western Cape,

RSA).

78

WATER OUTLET

WATER INLET

CELL HOLDER

POWER SWITCH

POWER LIGHT

TYGON TUBING

Figure 3.3. Schematic representation of a modified Franz cell multiple-cell drive unit

3.2.3. Number of samples

In vitro release testing of CP from topical formulations was conducted using a modified Franz

glass diffusion cell system, containing housings for six cells (Figure 3.3). Although systems with

three cells have been used [126, 129], a minimum of six cells or samples is recommended when

characterizing release rate profiles of an API from semi-solid products [49]. The use of six

samples is recommended so as to minimize the variability associated with individual sample

application to the donor compartments of diffusion cell systems [108].

3.2.4. Sampling times

The FDA species that at least five sampling times must to be used over a suitable time period in

order to ensure that an adequate in vitro release profile is generated and can be used to determine

API release rates from semi-solid products [49]. Sampling times may be varied depending on the

formulation matrix, but in most cases samples can be taken over the course of one day, unless the

release of the drug from the formulation is extremely slow or a sustained or controlled-release

profile is expected [115].

It has been reported that there is a specific time window during which samples for release

experiments should be taken [113]. Ideally, samples should be removed at times when the

influence of the membrane and its associated stagnant layer disappears and before excessive drug

depletion from the semi-solid dosage form being tested has taken place [17]. Two main reasons

79

have been put forward in an attempt to explain the rationale behind withdrawing samples during

this time window.

Firstly, it has been reported that the migration of an API from a semi-solid matrix to a receptor

medium occurs as a series of successive diffusional steps [137, 138]. In the initial stages the

membrane, together with an unstirred layer at its surface, have been reported to provide some

resistance to the permeation of API and thus affect the release rate of an API [113]. However,

over time, a diffusional layer is formed within the semi-solid matrix and thus becomes the rate-

controlling element [113]. As a consequence, samples should be withdrawn when the influence

of the membrane and its associated stagnant layer has disappeared [113], or otherwise the early

sampling time points should be ignored when calculating API release rates derived from such

systems [113].

Secondly, according to theoretical diffusion models a plot of the amount of API released vs. the

square root of time should be linear if drug release from a semi-solid matrix is diffusion

controlled or diffusion is the rate controlling element within the dosage form [137, 138].

However, it has been shown that API release from semi-solid dosage forms tends to deviate from

linearity at extended times [113] and the deviation is usually observed when more than

approximately 35-45% of the API in the dosage form has been released from the semi-solid

sample placed in the donor chamber [113]. Therefore, when choosing the upper time limit for

sample collection, this factor must be taken into consideration [113].

Preliminary in vitro release studies were undertaken over a three-day period using a

commercially available CP cream product viz., Dermovate® cream (Glaxo Wellcome SA (Pty)

Ltd, Midrand, Gauteng, RSA), a 0.1 µm Polycarbonate membrane filter (Millipore Co., Bedford,

MA, USA) and a receptor phase consisting of 50% v/v propylene glycol/water mixture (Section

3.2.6.2.3.3 and Section 3.2.6.3).

The data generated from these studies are shown in Figure 3.4. These data reveal that

approximately 3 % of CP in the semi-solid sample applied to the membrane was released after 2

hours. Therefore, it appears that two (2) hours was an adequate time for the influence of the

80

membrane and its associated stagnant layer to disappear, as the plot does not appear to show a

significant lag phase. Approximately 32% of CP contained in the formulation applied to the

membrane was released at the end of the 72-hour test period and this value is below the 35-45%,

after which non-linearity of release was expected to occur [17].

0

5

10

15

20

25

30

35

40

0 2 4 6 8 10

Time (h)0.5

Cum

ulat

ive

% r

elea

sed

Figure 3.4. In vitro release of CP from Dermovate® cream (n = 6)

Least squares linear regression analysis was used to determine the linearity of the in vitro release

profile for CP during the time period of 2 to 72 hours and the resultant coefficient of

determination (R2) of 0.9962, shown in Figure 3.5, reveals that a plot of % CP released vs. square

root of time was linear. In light of the constraints relating to the monitoring of API release from

semi-solid dosage forms, samples of the receptor medium were removed at 2, 4, 8, 12, 24, 48

and 72 hours in order to generate satisfactory CP release profiles and to characterize the

mechanism and type of release of CP from semi-solid dosage forms.

81

y = 4.2173x - 2.1687R2 = 0.9962

0

5

10

15

20

25

30

35

40

0 2 4 6 8 10

Time (h)0.5

Cum

ulat

ive

% r

elea

sed

Figure 3.5. Least squares linear regression best fit line of the in vitro release profile of CP from Dermovate® cream (n = 6)

3.2.5. Temperature

The temperature of the receptor medium is generally set to 32 ± 0.5˚C in an attempt to

approximate the natural surface temperature of the skin [102, 107, 111, 113, 115, 122, 123, 126].

An increase in temperature normally results in an increase in drug release rate and hence some

deviation from using 32˚C, for example the use of 37˚± 0.5˚C may be justifiable. However the

use of excessively high temperatures may melt the base or the product being tested or cause

significant physical changes to such a product, which in turn may change the resistance of the

matrix to drug diffusion into the receptor medium and should therefore, be avoided [113].

In vitro release experiments conducted at 32°C and 37˚C showed no significant differences in the

in vitro release rates of triamcinolone acetonide from various commercial cream and ointment

formulations [106]. Therefore the in vitro release profile characterization of CP release from

semi-solid formulations manufactured in these studies was conducted at 32° ± 0.5˚C.

82

3.2.6. Receptor medium

3.2.6.1. Overview

One of the most important considerations when developing an in vitro release test method is the

selection of a receptor medium that has adequate sink characteristics throughout the course of an

experiment [115]. In order to achieve favourable sink conditions and to ensure that the lower

surface of the membrane remained in contact with the receptor phase over the entire

experimental time period, each diffusion cell was completely emptied and refilled immediately

with fresh receptor fluid at each sampling time.

However, it is also worth noting that a receptor medium must have a high capacity to dissolve

the active ingredient that has been released from the formulation being tested [112]. This may be

accomplished by maintaining the thermodynamic activity of an API in a receptor fluid at less

than 10% of its thermodynamic activity in a donor medium, thereby maintaining a favourable

driving force for permeation and ensuring reasonable and efficient transfer and collection of the

permeant of interest [122]. Therefore, another important factor to consider when selecting a

receptor fluid is the solubility of an API being tested in a receptor fluid or medium [108]. It is

therefore critical that a drug substance has sufficient solubility in a receptor medium without

impacting on the sink conditions of that system [121].

Ideally, receptor media should be aqueous systems [115] and for most studies isotonic solutions,

buffered to a pH of 7.4 are preferred for use as receptor fluids [122]. However, for products

formulated with water-insoluble drugs, the selection of an appropriate receptor medium to

maintain sink characteristics is a challenge [121], and in order to facilitate and monitor drug

release from such topical formulations it may be necessary to alter receptor fluid pH, add

surfactants and/or complexing agents such as cyclodextrins [115] or use non-aqueous media in

which the drug is more soluble to efficiently dissolve the API that has been released from the

matrix during release studies [122].

83

3.2.6.2. Selection of receptor medium

3.2.6.2.1. Aqueous systems

CP is a hydrophobic corticosteroid with a limited aqueous solubility (Chapter 1, Section 1.3.1.1).

Consequently, the use of a purely aqueous receptor fluid was not considered. In addition, since

the drug is non-ionisable (Chapter 1, Section 1.3.2) altering the pH of an aqueous receptor

solution would not have an effect on its aqueous solubility and therefore the use of other solvent

systems was considered appropriate.

3.2.6.2.2. Water-immiscible systems

The hydrophobic nature of CP and the associated difficulties in selecting an appropriate receptor

medium might persuade researchers to use non-aqueous solvents such as acetone, chloroform or

dichloromethane in which the drug would be freely soluble as the receptor medium (Chapter 1,

Section 1.3.1.1). However, the use of such media in release studies is undesirable since these

solvents are likely to interfere with the analytical method of choice, especially if the analytical

procedure requires direct injection of the receptor medium onto the HPLC system as was desired

in these studies.

3.2.6.2.3. Water-miscible systems

3.2.6.2.3.1. Overview

A further practical consideration in the selection of an appropriate receptor medium is to

consider the use of water-miscible solvents such as ethanol and propylene glycol (PG) or

propane-1,2-diol. Ethanol, PG and aqueous solutions of various concentrations are widely used

as solvents in pharmaceutical formulations and cosmetics [139, 140]. Ethanol is a clear,

colourless, mobile and volatile liquid with a slight characteristic odour and burning taste [139]

and PG is a clear, colourless, viscous, practically odourless solvent with a sweet, slightly acrid

taste resembling that of glycerine [140].

84

Ethanol has a specific gravity or density of 0.8119-0.8139 g/cm3 at 20˚C [139] and the specific

gravity of PG is 1.038 g/cm3 at 20˚C [140]. These materials are generally regarded as relatively

non-toxic and chemically stable when stored in well-closed containers at cool temperatures [139,

140]. Since CP is relatively soluble in ethanol and PG (Sections 1.3.1.1 and 1.3.1.2.1, Chapter 1),

the use of ethanol and PG aqueous mixtures as potential receptor fluids for CP was considered in

preliminary studies.

3.2.6.2.3.2. Alcohol/water mixtures

The use of hydro-alcoholic solutions as receptor media for lipophilic drugs such as triamcinolone

acetate [106, 126], betamethasone dipropionate [121] and rooperol tetra-acetate [123] has been

reported. Consequently, ethanol ABS (Protea Chemicals, Port Elizabeth, Eastern Cape, RSA) at

a concentration of 30% v/v was assessed as a receptor medium for CP, using Silatos™ silicone

sheeting, REF 7458 (Atos Medical, Hőrby, Sweden) as a supporting membrane.

The data generated from these studies are illustrated in Figure 3.6., and are reported as the

cumulative amount of CP released per unit area (Q) vs. time (t). However the use of aqueous

ethanol solutions as a receptor medium for CP were not considered ideal because of the presence

of excessive air bubbles that were observed beneath the synthetic support membrane. Air bubbles

have been reported to interfere with the contact between the receptor medium and the supporting

membrane that is required, resulting in a reduced surface area available for drug diffusion to

occur [121].

85

0

1

2

3

4

5

6

7

8

9

10

0 20 40 60 80

Time (h)

Q (µ

g/cm

2 )

Figure 3.6. In vitro release profile of CP from Dermovate® cream, using 30% v/v ethanol solution as receptor medium (n = 6)

Visual inspection of the Franz cells at the end of the experiment also revealed noticeable

evaporation of the receptor medium, which resulted in a decrease in the volume of the receptor

fluid in some cells. The extent of evaporation of the receptor medium might have contributed to

the large standard deviations observed at the longer sample times such as 48 and 72 hours.

3.2.6.2.3.3. Propylene glycol/water mixtures

The use of PG as a receptor fluid has also been reported [141]. Therefore in preliminary studies

the in vitro release of CP through Silatos™ silicone sheeting, REF 7458 (Atos Medical, Hőrby,

Sweden) into a 30% v/v PG:water receptor medium was evaluated and the data from these

studies are shown in Figure 3.7. Unlike aqueous ethanol, the use of 30% v/v PG did not lead to

the formation of excessive of air bubbles beneath the surface of the membrane used in these

studies. Inspection of the Franz cells at the end of the experiment also revealed little evidence of

86

receptor medium evaporation and therefore this solution was considered promising as an

appropriate receptor medium for CP release studies.

0

2

4

6

8

10

12

14

0 20 40 60 80

Time (h)

Q (µ

g/cm

2 )

Figure 3.7. In vitro release profile of CP from Dermovate ® cream using 30% v/v PG solution as receptor medium (n = 6)

3.2.6.3. Saturation solubility

Following the selection of 30% v/v PG:70% v/v water as the most promising receptor medium

for CP release studies, saturation solubility studies of CP in water, PG and PG:water binary

mixtures were conducted in order to determine the optimal concentration of PG for use in a

receptor medium for CP.

The saturation solubility of CP was determined in 0, 30, 40, 50, 60 and 100% v/v PG.

Approximately 50 mg of CP was accurately weighed and placed into a test tube and 2.0 ml of the

different solvents to be tested were added. Samples were agitated at 120 oscillations per minute

87

in a Model-123 Labotec Oscillating water bath (Labotec (Pty) Ltd, Johannesburg, Gauteng,

RSA), maintained at 32○C for 48 hours.

The solutions were then filtered through a 0.22 µm Millipore filter (Millipore Co., Bedford, MA,

USA) and a 1.0 ml aliquot of the filtrate was diluted to 10 ml with acetonitrile and analysed

using a specific and sensitive, previously validated HPLC method with UV detection at 240 nm

(Chapter 2, vide infra).

A list of the experimentally determined saturation solubility values for CP in various solutions of

different concentrations of PG concentrations is listed in Table 3.1, and the saturation solubility

profile of CP in the various solutions, plotted as log solubility vs. PG concentration, is depicted

in Figure 3.8.

Table 3.1. Solubility data for CP in various concentrations of PG at 32˚C (n = 3)

PG concentration (%v/v) Mean saturation solubility (mg/ml)

SD

0 0.003600 0.0009000 30 0.01760 0.001680 40 0.05880 0.0007800 50 0.1983 0.0002900 60 1.334 0.2300 100 8.550 2.500

88

0.001

0.01

0.1

1

10

100

0 30 40 50 60 100

PG concentration (% v/v)

Log

Sol

ubili

ty (m

g/m

l)

Figure 3.8. Saturation solubility profile of CP in PG:water solutions of different proportions (n = 3)

The resultant data show that CP is poorly soluble in water, but relatively soluble in PG. It is clear

that as the PG content of the dissolution or receptor medium is increased, the saturation solubility

of CP also increases. The objective of conducting these studies was to determine the saturation

solubility of CP in a PG concentration that was greater than ten (10) times that of the maximum

achievable concentration that may be attained during the course of in vitro release experiments.

In the pilot study conducted and reported in Section 3.2.6.2.3.3 (Figure 3.7), approximately 300

mg of 0.05% w/w Dermovate® cream was applied to the synthetic membrane and the volume of

the receptor medium used was 12.5ml. Assuming that CP is freely soluble in the receptor

medium, the theoretical maximum concentration that would be achieved during the course of an

in vitro release study is calculated to be 12 µg/ml.

The saturation solubility data summarized in Table 3.1 reveal that CP has a saturation solubility

of 198.27 ± 0.29 µg/ml in a 50% v/v PG:water solution, corresponding to nearly 16.5 times the

estimated maximum concentration of 12 µg/ml that could be achieved. Therefore based on the

89

saturation solubility of CP in this solvent system and the fact that the receptor medium was

replaced at each sampling time during the course of the experiment, it was considered that

adequate sink conditions would prevail for the duration of all studies. Therefore a 50% v/v

PG:water solution was selected as an appropriate receptor phase for use in future in vitro release

studies of CP creams.

3.2.6.4. Preparation of the receptor medium

The receptor medium was prepared by carefully adding equal parts by volume of propane-1,2-

diol (Merck Chemicals (Pty) Ltd, Darmstadt, Germany) and HPLC-grade water generated from a

Milli-RO® 15 water purification system (Millipore Co., Bedford, MA, USA), to a Schott Duran®

glass solvent mixing bottle (Schott Duran GmbH, Hattenbergstrasse, Germany) using a

measuring cylinder. Following mixing, the mixture was allowed to equilibrate to room

temperature and then the receptor phase was degassed under vacuum with the aid of a Model A-

2S Eyela Aspirator (Rikakikai Co., Ltd, Tokyo, Japan) and filtered through a 0.45 µm Millipore®

HVLP filter (Millipore Co., Bedford, MA, USA) prior to use. The receptor phase was degassed

in order to remove dissolved air, thereby minimising the potential formation of air bubbles

beneath the supporting membrane, which have been reported to adversely affect the in vitro

release of an API from semi-solid dosage forms [122].

3.2.7. Synthetic membranes

3.2.7.1. Overview

Synthetic membranes have been used extensively to determine the in vitro release characteristics

of an API from topical formulations [102, 115, 117, 121, 134, 142]. These membranes are not

intended to mimic the barrier properties or the heterogeneous nature of the skin [113]. Instead,

they are designed to provide a physical support and maintain constant contact between the

formulation and dissolution medium and to prevent bulk transfer of the dosage form, whilst

allowing for the monitoring of API release from a formulation into a receptor medium [113,

115].

90

It has been reported that it is possible to study the in vitro release of a drug from rigid topical

formulations such as ointments into an aqueous medium in the absence of a separating

membrane [142]. However, creams and gels invariably contain an aqueous phase and/or adjuvant

components that are aqueous in nature, are water-miscible or water-soluble. Therefore a

membrane must be placed between the formulation being tested and the receptor medium in

order to maintain the physical integrity of individual formulations and/or dosage forms [112].

3.2.7.2. Characteristics of membranes

Synthetic membranes that have been commonly used for in vitro release studies are those with

porous characteristics such as cellulose acetate [102, 103, 107, 134], nitrocellulose [105, 121]

and polycarbonate membrane filters [104, 105 123], or alternatively homogeneous permeable

polymers such as silicone based filter membranes [102, 107, 129, 134].

Important considerations that must be taken into account when selecting a synthetic membrane

for use in in vitro drug release experiments include the requirement for the membrane to be

commercially available and have little or no capacity to bind the API of interest [112, 113]. In

addition, the membrane should have low reactivity with formulation components, be compatible

with the receptor medium and offer the least possible diffusional resistance to the permeant of

interest [112, 113, 119, 125]. In short, the membrane of choice should be inert and provide a

holding surface without barrier properties for the active ingredient and test formulation(s) [108].

The effects of several membranes on the in vitro release of CP from Dermovate® cream were

investigated. The membranes tested were a non-porous synthetic membrane, Silatos™ silicone

sheeting REF 7458 (Atos Medical, Hőrby, Sweden) and porous synthetic membranes such as

Isopore™ Membrane Filters (Millipore Co., Bedford, MA, USA) and MF-Millipore™ Membrane

Filters (Millipore Co., Bedford, MA, USA). The characteristics of the membranes used in these

studies are summarized in Table 3.2.

91

Table 3.2. Summary of the characteristics of synthetic membranes

Membrane Pore size (µm) Thickness (µm) Nature Silatos™ (Silicone) NA 120 Hydrophobic Isopore™ (Polycarbonate) 0.10 20 Hydrophilic Isopore™ (Polycarbonate) 0.40 20 Hydrophilic MF-Millipore™ (Nitrocellulose) 0.025 105 Hydrophilic MF-Millipore™ (Nitrocellulose) 0.45 150 Hydrophilic

3.2.7.3. Assessment of membranes

The effects on in vitro release of CP of each membrane were assessed using a modified Franz

diffusion cell system (Section 3.2.2) and a receptor medium comprised of 50% v/v PG:water

prepared as described in Section 3.2.6.3. The purpose of these studies was to determine the most

suitable membrane to be used as a support synthetic medium for the assessment of semi-solid

products containing 0.05% w/w CP. Each membrane was cut to approximately 30 mm in

diameter to fit the circumference of the membrane holder on the diffusion cell system and was

soaked in the receptor medium prior to use.

Cellulose membranes are believed to contain a number of water-soluble softeners, preservative

and plasticizer additives which may affect drug permeation [143]. To ensure the removal of these

substances, pieces of cellulose-based membranes were rinsed with distilled water prior to

soaking in the receptor fluid. After one hour, the membranes were carefully removed from the

receptor medium with the aid of a pair of tweezers, wiped with adsorbent tissue to remove excess

surface liquid and mounted between the donor and the receptor compartments of the Franz cell

assembly.

Parafilm ‘M’® Laboratory Film (Pechiney Plastic Packaging, Chicago, IL, USA) was used to

ensure that a leak-proof seal was formed between the flanges of the upper and lower components

of the diffusion cell assembly that was held together with a screw clamp. Following equilibration

of the cell to 32˚C an infinite dose of approximately 300 mg of the test formulation was applied

evenly to the entire surface of the membrane with the aid of a glass rod and the final dose was

determined by weighing the rod before and after application. The donor compartment was

covered with Parafilm® in order to achieve occlusive conditions throughout the experiment.

92

The receptor chamber of each diffusion cell was filled with 12.5 ml of the receptor medium and

stirred by means of a 2x2 mm star-head magnetic stirrer (Merck Chemicals (Pty) Ltd, Darmstadt,

Germany). The receptor compartment was examined for the presence of air bubbles and if

present bubbles were removed by manually tapping or rotating the diffusion cell apparatus

assembly. Aliquots of receptor medium were removed at the appropriate sampling times as

discussed in Section 3.2.4.

Following the removal of a sample to be analyzed, each diffusion cell was completely emptied

and refilled immediately with fresh receptor fluid. Samples were stored at room temperature

(22˚C) until analyzed. The samples were analyzed immediately after the conclusion of the

release studies i.e. within three days (72 hours) of the commencement of the study. These

samples were found to be stable at these storage conditions as investigated and reported in

Chapter 2, Section 2.5.2.6.3.2.

At the conclusion of release experiments, no physical changes to any of the membranes tested

were observed following exposure of the membrane to the PG-water receptor medium or the

semi-solid formulation applied to the membrane. These results were considered indicative of the

compatibility between the membranes and the receptor medium used, as well as the components

of the topical formulation containing CP. A specific, sensitive and validated HPLC method with

UV detection at 240 nm (Chapter 2) was used to analyze the samples and to determine the

amount of drug released per unit area. The cumulative amount of CP released (Q) over the 72

hour time period was plotted against time.

The results obtained following membrane selection studies are depicted in Figure 3.9. It is

immediately apparent that the in vitro release rate of CP through the silicone membrane was

significantly slower than through any of the nitrocellulose and polycarbonate membranes tested.

The highest amount of CP that was released was 56.37 ± 5.02 µg/cm2 and was released through

the 0.45 µm nitrocellulose membrane, whereas the lowest amount of 6.69 ± 1.11 µg/cm2 was

released through the silicone membrane. These data are discussed further in Section 3.2.7.4.

93

0

10

20

30

40

50

60

70

0 10 20 30 40 50 60 70 80

Time (h)

Q (µ

g/cm

2 )

Silicone 0.10 µm polycarbonate0.40 µm polycarbonate 0.025 µm nitrocellulose0.45 µm nitrocellulose

Figure 3.9. Effect of membrane type on the in vitro release of CP from Dermovate® cream (n = 3)

3.2.7.4. Membrane selection

The migration of a drug from a semi-solid matrix into a receptor medium is essentially a function

of one or a combination of the following three processes:

a) drug release from the semi-solid matrix itself,

b) passage of the drug through the membrane, and,

c) clearance of the drug from below the membrane.

In cases where neither of the last two processes are rate-limiting it is only the thermodynamic

activity of the drug in the semi-solid matrix that is an important determinant of drug release

[144]. Since the physical properties of the product itself are of interest, diffusion through the

base should be the rate-limiting step [113]. It is therefore critical that the membrane and the

receptor medium be highly permeable and accessible to the drug in the formulation to effect drug

release.

94

3.2.7.4.1. Silicone membrane

Silicone is a relatively inert, lipophilic and non-porous membrane as summarized in Table 3.2.

The lipophilic nature of silcone provides an ideal environment for partitioning and permeation of

lipophilic drugs, whilst its aporosity provides some rate-limiting function to this process [143].

Consequently, these membranes are useful for use as surrogate membranes for human skin when

evaluating in vitro release characteristics of a drug from topical or transdermal formulations.

The in vitro release profile of CP release generated through the silicone membrane suggested that

the membrane behaves as a rate-limiting component of the test system, since the passage of only

4.42 ± 1.11% of the applied dose into the receptor medium over the 72 hour test period was

allowed. This percentage release was considered low and appeared to indicate the lack of free

passage of CP through the membrane into the receptor medium.

The objective of these studies was to find a synthetic membrane with the least possible resistance

to CP diffusion. Therefore the use of a silicone membrane in this test system was considered

inappropriate for these studies as the use of this membrane would not provide relevant

information pertaining to CP release from test formulations but would inhibit release and provide

information pertaining to the flux of the drug molecule via membrane-limited release.

3.2.7.4.2. Porous membranes

Cellulose membranes have been used extensively in diffusion cell test systems for quality control

studies [102, 103, 105, 107, 121, 134]. It is worth noting that pure cellulose acetate membranes

are no longer commercially available, due to an Environmental Protection Agency (EPA) ban on

one of the chemicals needed to manufacture the membranes [102]. The membranes currently

referred to as cellulose acetate membranes contain a mixture of mixed esters of cellulose acetate

and cellulose nitrate, unlike the pure cellulose acetate membrane which contained cellulose

acetate only. Generally these membranes are reported to be more permeable than biological

membranes or aporous synthetic media and to allow the passage of diffusing molecule(s)

irrespective of the physicochemical characteristics of the compound under investigation [143].

95

Similarly porous polycarbonate filter membranes generally appear to be useful for use as a

dividing medium or as a supporting screen in test systems, where the release rate of a drug from

a delivery system is under investigation, as opposed to the evaluation of the kinetics of release of

a permeant from a formulation [143]. These filter media do not simulate the skin and provide no

significant barrier to the diffusion of a drug molecule but rather provide a relevant support for a

test formulation as is required for these types of studies.

The highest rate and amount of CP release was observed when using a 0.45 µm nitrocellulose

membrane, implying that it had the lowest resistance to the diffusion of CP. It would therefore

seem appropriate to select a 0.45 µm nitrocellulose membrane as the support membrane of

choice. However, a series of assumptions have to be made when selecting a membrane that

offers little resistance to drug penetration for use in in vitro release studies [138]. Among these

assumptions is the fact that there should be no more than 30% of the total amount of an applied

dose released into a receptor medium at the end of an experiment [138]. The cumulative

percentage CP released through each of the hydrophilic membranes studied over the 72 hour

period is summarized in Table 3.3.

Table 3.3. Cumulative percentage CP released after 72 hour (n = 3)

Membrane Pore size (µm) Mean % CP released/cm2

SD

Nitrocellulose 0.025 26.75 0.6100 Polycarbonate 0.40 29.53 2.720 Polycarbonate 0.10 32.53 1.970 Nitrocellulose 0.45 37.38 5.020

It is evident from the data summarized in Table 3.3 that the 0.45 µm nitrocellulose membrane

allowed for the passage of more than 30% of the drug placed in the donor chamber and therefore,

based on the assumption that less than 30% should be released as reported by Higuchi [138], this

membrane was deemed inappropriate for use. Similarly the 0.1 µm and 0.4 µm polycarbonate

membranes were also deemed inappropriate for the assessment of CP release from semi-solid

formulations. The cumulative percent CP released during in vitro release testing with a 0.025 µm

nitrocellulose was approximately 27% and therefore the 0.025 µm nitrocellulose membrane was

considered the membrane of choice for further studies (Section 3.2.7.5).

96

3.2.7.5. Membrane resistance

Following selection of the 0.025 µm nitrocellulose membrane as the most suitable synthetic

membrane for formulation development studies, membrane resistance to free permeation of CP

was evaluated by assessing the release of CP from a 0.05% w/v solution of CP made up in the

receptor medium (50% v/v PG: 50% v/v water). In addition 300 µl of the 0.05% w/v solution

was substituted for 300 mg of semi-solid formulation in the donor compartment. The release

profile generated from the CP solution was then compared to the profile generated following

testing CP release from Dermovate® cream.

The data generated in these studies are depicted in Figure 3.10. The results revealed that the in

vitro release of CP through a 0.025 µm nitrocellulose membrane was significantly slower from

the cream formulation than from the 0.05% w/v CP solution. Within the first 8 hours of the

release experiment, 4.82 ± 0.98% and 29.16 ± 1.92% of CP had been released from the cream

and solution respectively. A total of 24.4 ± 3.75% and 46.04 ± 2.07% of CP were released from

the cream and solution, respectively, at the end of the 72 hour experiment.

97

0

10

20

30

40

50

60

0 10 20 30 40 50 60 70 80

Time (h)

Cum

ulat

ive

% r

elea

sed

Dermovate® cream 0.05% w/w CP solution

Figure 3.10. Comparison of in vitro release of CP from a 0.05% v/v CP solution and Dermovate® cream using a 0.025 µm nitrocellulose membrane (n = 6)

The fact that the permeation of CP across the 0.025 µm nitrocellulose membrane was faster from

the CP solution than from the cream suggested that the membrane was not a rate-limiting factor

in diffusion and did not affect CP release from the cream formulation and test solution. The

differences observed in the release rate profiles for CP could therefore be directly attributed to

the characteristics of the semi-solid formulation and not the membrane used in the release rate

studies. Consequently the 0.025 µm nitrocellulose membrane was selected as an appropriate

inert, porous and commercially available synthetic membrane for the assessment of in vitro

release from topical formulations containing 0.05% w/w CP.

98

3.2.8. Sample Application

3.2.8.1. Overview

Depending on the method used to interpret the release data, either a finite or infinite dose may be

applied to the donor compartment of a test system [115]. The use of finite dosing results in the

release of the majority of the drug applied to the surface of a membrane into the receptor

medium, whereas with infinite dosing only a small percentage of the drug that is applied to the

membrane is released into the receptor medium over the course of the experiment [115].

Generally, data analysis is far less complicated when using infinite dosing or in situations in

which less than 30% of the applied dose is released during a test period [115]. Consequently the

effects of the amount of formulation applied to a membrane on the in vitro release rate of CP

were investigated. The objective was to determine an appropriate amount of formulation to be

applied to the membrane in order to correspond to an infinite dose condition or application.

3.2.8.2. Effects of sample application

The effect of applying different amounts of a sample to be tested into the donor compartment

was studied using a modified Franz glass diffusion cell system (Section 3.2.2.4), a 50% v/v

PG:water receptor medium (Section 3.2.6.3) and a 0.025 µm nitrocellulose membrane (Section

3.2.7.2). Approximately 50 mg, 100 mg, 200 mg and 300 mg of 0.05% Dermovate® cream was

accurately weighed and spread evenly over the entire surface of the membrane using a glass-

stirring rod. The glass rod was weighed before and after application and an accurate weight of

cream applied to the membrane was recorded. This technique was found to be reproducible with

respect to the sample weight applied to the membrane in all tests. The results obtained from these

studies are shown in Figure 3.11.

99

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2)

50 mg 100 mg 200 mg 300 mg

Figure 3.11. Effect of sample loading on the in vitro release of CP from Dermovate® cream

The results shown in Figure 3.11 are reported as the amount released per unit area (Q) vs. the

square root of time (t1/2) as opposed to time. In this manner the effect of sample amount on in

vitro release rates of CP may be explained using theoretical diffusion models. The data reveal

that the amount of drug released (Q) vs. t1/2 increased as the amount of semi-solid applied to the

membrane was increased. The Higuchi diffusion model [137, 144] which can be described using

Equation 3.1 may be used to describe the release characteristics of a drug from a suspension or in

solution in a semi-solid base.

( )[ ] 2/12 tDCCAQ sss−= Equation 3.1

Where:

Q = the cumulative amount of drug released,

Ds = the diffusivity of drug in the semi-solid matrix,

A = the amount of drug in the formulation, and

Cs = the saturation solubility of the drug in the semi-solid matrix.

t = time

100

The Higuchi equation predicts that a plot of Q vs. t1/2 would be linear for a drug in suspension in

a semi-solid matrix with an infinite dose application provided that such release is diffusion-

controlled by the matrix material that comprises the semi-solid dosage form [137].

Least squares linear regression analysis was used to determine the degree of linearity of the plots,

and the resultant coefficients of determination (R2) indicate that all the plots were linear over the

time period studied. The data are summarized in Table 3.4.

Table 3.4. In vitro release characteristics from four loading doses applied to the test membrane

The data shown in Table 3.3 imply that any of the amounts of semi-solid (50, 100, 200 and 300

mg) applied to the membrane in these studies conformed to Higuchi kinetics and although the

results suggest that an infinite dose condition was achieved with all four amounts of formulation

tested, 300 mg was selected for use in these studies due to the ease of experimental set-up, for

example weighing 300 mg by difference was considered much quicker and easier than weighing

smaller quantities.

3.2.9. Sample occlusion

3.2.9.1. Overview

The FDA guidance document [49] recommends that the amount of the semi-solid preparation

applied uniformly onto a membrane should be kept occluded during the assessment of the in

vitro release characteristics of a drug from that formulation. Occlusion is preferred in order to

prevent solvent evaporation and compositional changes of a formulation that is applied to the test

membrane [49].

The term occlusion refers to the use of a moderately impermeable barrier to cover the donor

compartment of a diffusion cell assembly [115]. Experiments were performed in the sample

Loading dose (mg) R2 50 0.9846

100 0.9896 200 0.9873 300 0.9881

101

occluded or non-occluded mode to determine whether the degree of occlusion affects the in vitro

release of CP and to select the appropriate conditions for future studies.

3.2.9.2. Effects of occlusion

The effect of occlusion on the in vitro release of CP from Dermovate® cream was investigated

using approximately 300 mg of the cream formulation. Diffusion cells were occluded by placing

a piece of Parafilm® over the donor compartment of the diffusion cell and sealing the cell.

The effects of occlusion and non-occlusion of the donor cell on the in vitro release characteristics

of CP from an infinite dose condition are shown in Figure 3.12. These data reveal that the

amount of CP released in the in vitro release profiles from an infinite dose condition under

occlusion was greater than from an un-occluded test system.

05

1015

2025

3035

4045

50

0 10 20 30 40 50 60 70 80

Time (h)

Q (µ

g/cm

2 )

Occluded Non-occluded

Figure 3.12. Effect of occlusion on the in vitro release of CP from Dermovate® cream (n = 6)

102

Many semi-solid formulations contain solvents such as ethanol, water and propylene glycol and

it has been reported that ethanol and water evaporate rapidly, whereas propylene glycol only

evaporates appreciably over a 24 hour period [112]. The composition of a semi-solid formulation

being tested would therefore change over the experimental period, should the test system not be

occluded for the duration of the experiment.

It is possible that the use of non-occluded cells resulted in the evaporation of water as well as

propylene glycol over 24 hours which might have led to an increase in the intrinsic viscosity of

the formulation, thereby causing an increased resistance to the diffusion of CP and a subsequent

lower overall release rate as shown in Figure 3.12. Consequently the in vitro release profile of

CP release from the topical formulation was markedly decreased and the deviation between the

two curves was more noticeable for the period 24-72 hours of the diffusion experiment.

As a result of the improved extent of release observed when using occluded cells, all future in

vitro release experiments for the assessment of CP release from semi-solid formulations were

carried out under occlusive conditions to prevent solvent evaporation and compositional changes

of the formulations being tested as such changes may influence the overall cumulative amount of

CP released during testing.

3.2.10. Sample Analysis

A sensitive, selective, accurate and precise previously validated HPLC method with UV

detection at 240 nm (Chapter 2) was used to analyze the samples withdrawn from the receptor

chambers at each time-point. The concentration of CP in each receptor cell was determined by

interpolation of the peak height ratios of CP to I.S. from a calibration curve. A typical calibration

curve is shown in Figure 3.13 and each calibration curve was prepared on the day on which

samples were analysed.

103

y = 0.1038x - 0.0041R2 = 0.9999

0.0

0.5

1.0

1.5

2.0

0 5 10 15 20

Conc. (µg/ml)

Peak

hei

ght r

atio

Figure 3.13. CP calibration curve used to determine the amount of CP released in in vitro release experiments

3.2.11. Comparison of diffusion or release rate profiles

The in vitro release profiles generated from two semi-solid products (test and reference) were

compared using a statistical test recommended by the FDA to assess product sameness when

changes are made to an approved topical semi-solid product [49]. The analysis involves non-

parametric assessment of release rate profiles by statistical means and is based on the calculation

of an appropriate confidence interval (C.I.).

The test is related to the Wilcoxon Rank Sum/Mann-Whitney Rank Test that is applied to the log

of the slope of the release profiles or the release rate. The in vitro release profiles of two

formulations are generated for a test and a reference (T and R) product (n = 6) and the resulting

slopes of the profiles or estimated in vitro release rates for each of the formulations are

calculated. A 90% C.I. for the ratio of the median in vitro release rate or population for the test

104

product over the in vitro release rate or population for the reference product is computed and

expressed in percentage terms.

The computation of the C.I. is achieved by evaluation of 36 (6 x 6) individual ratios of T/R in

vitro release rates and are ordered in sequence, i.e. all possible T/R ratios (36 in total) are

arranged in ascending order from lowest to highest with respect to relative release rates. The

eighth (8th) and twenty-ninth (29th) individual ratios are then established and are assigned as the

lower and upper limits of the 90% C.I. for the T/R release rate ratio respectively and converted

into percentages. In order to accept that the in vitro release rate of a test product (T) is within the

90% C.I. of a reference product (R), the values for T/R ratios should fall within the limits of a

0.75 or 75% to 1.33 or 133% C.I.

An example of the calculation of the confidence interval is shown in Appendix 1. The data

shown were generated in in vitro release studies in which CP release from an extemporaneously

manufactured CP cream (T) and from Dermovate® cream (R) were assessed (Chapter 4, vide

infra). The eighth and twenty-ninth ordered individual T/R ratios are 101.6% and 125.3%, which

fall within the 75% -133% limits. Therefore the release rates from the two semi-solid

formulations in this example are considered identical and as a result the two formulations can be

considered equivalent.

3.2.12. Optimal in vitro release test conditions

The optimal in vitro release test conditions established for the assessment of CP release from

topical formulations are summarised in Table 3.5.

105

Table 3.5. Summary of optimal in vitro release test conditions

Parameter Condition Apparatus Sample number Average diffusional surface area Average receptor volume Receptor medium Temperature Synthetic membrane Dosing conditions Sample occlusion/non-occlusion Magnetic stirrers Sampling time Sample analysis Comparison of diffusion profiles

Vertical glass Franz cell diffusion cell 6 cells 1.921 ± 0.056 cm2 12.50 ± 0.38 ml 50% propylene glycol:50% water 32 ± 0.5˚C. 0.025 µm nitrocellulose Infinite (300 mg sample application) Occlusion 2x2 mm star-headed magnetic stirrers 0, 2, 4, 8, 12, 24, 48, and 72 hours HPLC with UV detection at 240 nm Non-parametric statistical test

3.3. METHOD VALIDATION

3.3.1. Overview

The fundamental principle on which the validation of an in vitro release test method is based is

that following validation there is an assurance that the test method is performing as expected and

that when changes are made to the formulation composition, batches or sources of ingredients

and method of manufacture, changes in drug release rates can be detected [113, 115].

The integrity of this in vitro release method was therefore interrogated by investigating the effect

of changing various formulation attributes on the CP release rate from a prototype

extemporaneous cream. The variables validated in these studies were changes in:

a) dose strength,

b) composition, and,

c) viscosity of the semi-solid dosage form.

The use of these attributes as parameters in validation protocols for in vitro release test methods

have been previously reported [108, 113, 115].

106

The slope of the linear portion of the cumulative amount of drug released vs. the square root of

time (Q vs. t1/2) plot has been reported as the most logical, meaningful and accurate parameter for

comparing in vitro release characteristics of a drug from different topical formulations [113].

Thakker and Wendy [108] used flux values in the validation of a typical in vitro release test

method and reported flux values as the average slope of the best fit line of a Q vs. t1/2 plot [108].

The approach adopted by Thakker and Wendy [108] was used to calculate flux values for CP

release from the test product and flux values were used to confirm the ability of the in vitro

release test method to differentiate between semi-solid formulations of CP that differed in

strength, excipient composition and viscosity.

3.3.2. Changes in dosage strength

The effects of changes in strength of CP in the formulation on in vitro release rate of CP were

evaluated using extemporaneously manufactured cream formulations at three different

concentrations levels, specifically, 0.01, 0.025 and 0.05% w/w of CP.

According to theoretical principles, the release rate and ultimate amount of drug released from a

specific formulation should be proportional to the amount of drug in the formulation if the

diffusion of the drug within the semi-solid matrix is the rate-controlling mechanism in drug

release [138, 144]. Data generated in these studies are summarized in Table 3.6 and shown in

Figure 3.14 and are in agreement with the theory. The higher the concentration of CP in the

semi-solid formulation, the greater the flux and ultimately the total cumulative amount of CP

release per unit area.

Table 3.6. Effect of changes in the CP concentration on the total cumulative amount released and the associated flux values (n = 3)

Dosage strength (%w/w) Q (µg/cm2) Flux (µg/cm2/h1/2) 0.010 11.22 1.470 0.025 26.09 3.540 0.050 41.81 5.410

107

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )

0.01% w/w 0.025% w/w 0.05% w/w

Figure 3.14. Effect of changes in CP content on the in vitro release rate of CP from an extemperaneous cream base (n = 3)

The data suggest that the selected receptor phase, support membrane and other experimental

conditions are suitable to show discrimination between different formulations since the rate-

controlling mechanism for drug release is diffusion of CP within the semi-solid base and not

through the membrane used in the cell assembly. Consequently, the impact of formulation

changes that may result in drug loading differences can be assessed using this in vitro release

method.

3.3.3. Changes in composition

The effect of changes in formulation composition on the in vitro release rate of CP was examined

using two extemporaneously manufactured creams formulations (Formulations 1 and 2). Both

formulations were manufactured using the same method of manufacture and contained the same

amount of CP (0.05% w/w). The difference between the two formulations was that a different

108

primary emulsifier was used in each case. Formulation 1 contained 1% w/w Ritapro® 200 (Rita,

Crystal Lake, IL, USA) and Formulation 2 contained 1% w/w Emulcire® 61 WL (Gattefossé

SAS, Saint-Priest Cedex, France) as the primary emulsifier. The data generated from these

studies are summarized in Table 3.7 and shown in Figure 3.15 and reveal that alterations in the

composition of the CP formulation do in fact result in changes in flux in addition to the total

cumulative amount of CP released. These data confirm the ability of this in vitro release method

to detect the impact of changes of formulation constituents on the in vitro release of CP from

such formulations.

Table 3.7. Effect of changes in formulation composition on flux and cumulative amount of CP released (n = 3)

CP cream Formulation

Dose strength (%w/w)

Primary Emulsifier

Q (µg/cm2)

Flux (µg/cm2/h1/2)

1 0.05 Ritapro® 200 63.06 ± 4.84 8.63 2 0.05 Emulcire® 44.47 ± 3.63 6.14

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )

Formulation 1 (Ritapro® 200) Formulation 2 (Emulcire® 61 WL)

Figure 3.15. Effect of changes in formulation composition on CP in vitro release rate (n = 3).

109

3.3.2.3. Changes in viscosity

3.3.2.3.1. Overview

In order for a drug molecule to be released across a membrane from a semi-solid vehicle, it has

to reach the membrane surface prior to partitioning and mass transport of the drug from the

formulation. Mass transport occurs by diffusion of drug molecules through the formulation base

[145] and API diffusion depends on the properties of the drug, the diffusion medium and the

extent of vehicle-drug interactions that may be present [146]. The extent and degree of these

interactions will in turn determine the rate and extent of drug release and the overall shape of the

drug release profile.

It is well established that the diffusivity of a drug molecule through a semi-solid material

decreases progressively as the viscosity of the base material increases [146]. The converse is also

true and the implication is that any alteration in the intrinsic viscosity of a topical formulation

would have an effect on the rate and extent of release of an active ingredient from a vehicle. It

follows therefore that the intrinsic viscosity of a formulation is a key attribute of semi-solid

dosage forms and a parameter that can be monitored for quality control purposes during and after

formulation development.

The effect of changes in the intrinsic viscosity of the semi-solid formulation base on the in vitro

release of CP was therefore investigated using formulations of different viscosity. Two prototype

cream formulations (Formulations A and B) were manufactured extemporaneously using similar

exicipients and method of manufacture. The formulation compositions differed in the amount of

consistency modifier, viz., white beeswax that was added to each formulation. Formulation A

was manufactured using 0.6% w/w white beeswax whereas formulation B contained 2% w/w

white beeswax. The exact intrinsic viscosity of each formulation was ascertained as described in

Section 3.3.2.3.2, vide infra.

110

3.3.2.3.2. Determination of viscosity

The intrinsic viscosity of Formulations A and B were measured using a Model-RVDI+

Brookfield Viscometer (Brookfield ENG Labs Inc., Stoughton, MA, USA). The viscometer was

operated at 100 rpm using a T-F (code 96) spindle and a Helipath stand and the spindle was

chosen to maintain a torque between 10% and 90%. In order to obtain stable display readings, all

measurements were recorded 60 sec after the commencement of spindle rotation and a maximum

of three (3) readings were taken to obtain an average viscosity value.

The intrinsic viscosity readings obtained were 14.83 ± 0.05773 KcP and 21.37 ± 0.1071 KcP for

Formulations A and B, respectively. These results indicate that Formulation B has a higher

intrinsic viscosity than Formulation A. In vitro release studies were conducted as previously

described in order to ascertain whether the intrinsic viscosity resulted in a different flux for CP in

addition to total cumulative amount CP released.

3.3.2.3.3. Effects of viscosity

The effects of viscosity of the semi-solid formulation on the in vitro release of CP are tabulated

in Table 3.8 and shown in Figure 3.16.

Table 3.8. Cumulative amount of CP released and the average in vitro release rate (flux) from CP cream formulations of different intrinsic viscosity (n = 3)

CP cream Formulation

Dose strength (%w/w)

Viscosity (KcP)

Q (µg/cm2)

Flux (µg/cm2/h1/2)

A 0.05 14.83 ± 0.05773 41.81 ± 0.54 5.41 B 0.05 21.37 ± 0.1071 36.21 ± 1.71 4.95

111

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )

Formulation A (14.83 KcP) Formulation B (21.37 KcP)

Figure 3.16. Effect of changes in the intrinsic viscosity of a formulation on the in vitro release rate of CP (n = 3).

These data indicate that there was a decrease in CP flux as well as the total cumulative amount of

CP released from the semi-solid formulation of higher viscosity. However, the CP release rates

from Formulation A and Formulation B, respectively, appear to be similar, probably more than

likely due to the small difference in the intrinsic viscosity values obtained for these two

formulations. Nevertheless, the results show that this in vitro release method is suitable and

capable of detecting the impact of small changes of the physical properties of CP semi-solid

formulations on the rate and extent of CP release from these products.

112

3.4. CONCLUSIONS

An in vitro release method for the assessment of clobetasol 17-propionate (CP) release from

topical semi-solid dosage forms has been developed and validated in terms of protocols reported

[108, 113, 115]. Method development studies entailed the selection of a suitable diffusion cell

system, appropriate sampling times, receptor medium and appropriate synthetic membranes in

addition to investigating the effects of sample application and sample occlusion or non-occlusion

on the in vitro release of CP from semi-solid dosage forms.

A modified Franz glass diffusion cell system (n = 6) was selected and used in these studies since

it has the most potential for use as a standardized test system that may be adapted for use as a

compendial method. Various solvent systems and compositions were investigated for use as

receptor fluids during the course of these experiments and a 50% propylene glycol: 50% water

mixture was deemed to be the optimal receptor medium as it was shown to have appropriate sink

characteristics for CP following saturation solubility studies.

There are certain constraints that are associated with the selection of sampling times and these

must be considered and investigated when monitoring the rate and extent of API release from

semi-solid dosage forms. Samples of the receptor medium were withdrawn at 2, 4, 8, 12, 24, 48

and 72 hours in order to generate a satisfactory CP release profile and to characterize the release

of CP from semi-solid topical formulations.

Various synthetic membranes were evaluated as potential physical barriers between the semi-

solid sample being tested in the donor chamber and the receptor medium in the receptor chamber

of the modified Franz cell assembly. A 0.025 µm nitrocellulose membrane was found to be the

most appropriate, inert, porous and commercially available synthetic membrane for the

assessment of in vitro release of CP from topical formulations. This synthetic membrane was

shown to have no resistance to the free diffusion of CP when tested using a solution of CP in a

propylene glycol water mixture.

113

The temperature of the test system was set at 32° ± 0.5˚C to approximate the usual surface

temperature of the skin and a 2 x 2 mm star head magnetic stirrer was used to agitate the receptor

medium. About 300 mg of the semi-solid preparation, which corresponded to an infinite dose,

was applied uniformly to the membrane, and the formulation was occluded for the duration of

testing, to prevent solvent evaporation and compositional changes to the formulation.

At the conclusion of studies to define the optimal parameters for in vitro release testing, the

method was assessed for its ability to detect the effects of changes in formulation characteristics

on CP release. The effects of changes of strength, composition and intrinsic viscosity of topical

formulations on in vitro release of CP were then evaluated. The results obtained indicated that

the in vitro release method that has been developed is able to detect the effects of changes in

formulation in which CP is suspended and/or dissolved on CP release and was therefore applied

in all future formulation development studies.

It is evident, from the data presented that the in vitro release test method is simple, reliable,

reproducible and discriminatory. It can therefore be used in conjunction with traditional quality

control tests to determine the quality and consistency of extemporaneously manufactured CP

topical formulations in addition to the assessment of the associated rate and extent of release of

the API during product development studies.

114

CHAPTER FOUR

DEVELOPMENT AND IN VITRO CHARACTERIZATION OF CLOBETASOL 17-

PROPIONATE CREAM FORMULATIONS

4.1. INTRODUCTION

The elevated costs of brand name or innovator pharmaceutical products has been reported as one

of the major contributing factors to the rapidly escalating costs of health care services worldwide

[147-150]. As a consequence, there has been a growing international trend for the demand of

safe, effective and, more importantly, affordable medicines [151]. A number of mechanisms

have been evaluated in an attempt to make pharmaceutical products more affordable and more

accessible, especially in the developing world where lifesaving medicines are reported to be

financially out of reach of the wider population [152-154].

In the Republic of South Africa (RSA) for example, generic prescribing and generic substitution

have been identified and targeted as some of the possible strategies to contain the escalating

costs of brand name pharmaceutical products, and these are currently being implemented [147,

149, 155, 156]. Generic prescribing refers to the prescription of multi-source (generic) medicines

by authorized prescribers, such as for example, medical doctors [155], whereas generic

substitution refers to the replacement of an innovator product with a multi-source (generic)

product by pharmacists [155].

Generic medicinal products have been described as products that are equivalent to brand name

products in terms of the strength of active pharmaceutical ingredients (API), dosage form, route

of administration, quality, safety, efficacy, performance characteristics and therapeutic indication

[151, 157]. However, generic products are typically sold at much lower prices than brand name

products and therefore are more affordable than innovator products [151]. It has been reported

that the low cost of generic pharmaceutical products is usually attributed to the fact that an

Abbreviated New Drug Application (ANDA) process [151, 157] does not require the generic

product manufacturer or sponsor to repeat costly and time-consuming clinical and non-clinical

115

studies on an API or dosage form (s) in order to demonstrate the safety and efficacy of the drug

or drug product(s) in question [151, 157].

Sponsors of generic drug products are required to satisfy specific criteria for the registration of

generic drug products as set by relevant regulatory authorities such as, for example, the office of

generic drugs of the Food and Drug Administration in the United States (FDA) [157], the

European Agency for the Evaluation of Medicinal Products (EMEA) [158] and the Medicines

Control Council (MCC) in South Africa [159].

It has been argued that the time, risk and cost associated with the development of brand name

pharmaceutical products are responsible for the high costs of innovator medicinal products [160].

According to a report compiled by the European Federation of Pharmaceutical Industries and

Associations (EFPIA) [161], an average of between 12-13 years is required for a medicinal

product to reach pharmacy shelves from discovery, costing an estimated € 870 million during the

research and development (R&D) phase for a product [161]. Furthermore, the EFPIA report

[161] alludes to the fact that only one (1) or two (2) out of ten thousand (10000) potential drug

substances synthesized in laboratories are able to pass the extensive testing in the R&D stages of

drug development to produce a marketable pharmaceutical product [161].

As a consequence, innovator medicinal products are developed under patent protection, which

protects the investment in the development of the drug product by giving the innovator company

the sole right to sell the product while the patent is in effect [151, 157]. As a matter of interest,

innovator pharmaceutical companies may apply and obtain patent rights from the United States

Patent and Trademark Office at anytime during the lifetime of the drug or drug product [151].

When the intellectual property rights for a product have been exhausted, normally after a period

of twenty (20) years from the date of filing for the patent [151, 160] generic medicinal product

manufacturers may then apply to the relevant authorities, such as the FDA for permission to

produce and sell a generic version of the branded product [151, 157].

Shargel and Kanfer [151] discussed various aspects that should be considered when selecting a

generic drug product for manufacture. They [151] argue that the estimated sales volume for the

116

innovator product and the potential market share the manufacturer of the generic product expects

to achieve once the multi-source product is manufactured and approved for marketing is the main

driving force for the selection of drug candidates for generic product manufacture [151].

Shargel and Kanfer [151] also argue that patent expiration and exclusivity as well as other legal

issues must be considered carefully prior to selecting an innovator drug and/or drug product for

generic manufacture. The manufacturers of a generic drug product will continue to face various

legal and patent challenges from innovator companies, which in the opinion of Shargel and

Kanfer hinder the entry of generic products into the marketplace [151].

Eczema or dermatitis is a common dermatological disorder affecting approximately one-third of

a given population [162] and may be defined as superficial inflammation of the skin, that is

characterized histologically by epidermal oedema and clinically by vesicles, poorly marginated

redness, oedema, oozing, crusting, scaling, pruritis and lichenification and that is usually caused

by scratching or rubbing of the skin [163]. Hartshorne [162] reported that eczema is the most

common skin condition in the RSA accounting for approximately 34.5%, 32.7%, 30.4% and

17.8% of the dermatological outpatients in the coloured, black, Indian and white populations,

respectively. Topical corticosteroid formulations such as creams or ointments applied three (3)

times daily have been reported to be the primary therapy for the treatment of eczema [163, 164].

Topical corticosteroids are, however, reportedly considered expensive for the majority of the

population, and as a consequence the supplemental use of other substances such as white

petrolatum and hydrogenated vegetable oil has been reported [163]. The high costs of therapy are

especially a result of the use of super-potent topical corticosteroids such as clobetasol 17-

propionate (CP) that are required for the treatment of severe or chronic eczema, especially of the

hands and feet (Section 1.5.2, Chapter 1). The elevated costs of super-potent topical

corticosteroid products, such as CP, especially in developing countries such as the RSA, may be

ascribed to the lack of a variety of generic versions of the product on the local market. For

example, despite the expiration of the patent for CP in 1990 [165], only one generic version of

Dermovate® cream and Dermovate® ointment (GlaxoSmithKline plc, Brentford, Middlesex,

U.K) has been registered for marketing in the RSA [166].

117

The objective of these studies was therefore to design and develop a generic version of

Dermovate® cream and to evaluate the product in terms of a number of in vitro performance

characteristics. The successful development and subsequent determination of equivalence

between the generic product and the innovator formulation in terms of in vitro performance

characteristics would form the basis for future studies, such as the evaluation of bioequivalence

of the generic CP product relative to Dermovate® cream, potentially resulting in the development

of a marketable generic CP cream.

It has been reported that pharmaceutical manufacturers frequently manufacture topical

preparations of a drug in both cream and ointment vehicles to satisfy the preference of the patient

and dermatologist [167]. It has been argued that many patients and physicians prefer creams to

ointments, because the former are easier to apply and remove and may have a cooling sensation

on the skin [167-170]. Ointments are usually more difficult to apply or to remove, but are

expected to act as a barrier and increase the hydration of the skin [168].

The development of a generic version of Dermovate® cream rather than Dermovate® ointment

was considered most appropriate, since it was assumed that a generic version of Dermovate®

cream would have a larger market share than a generic Dermovate® ointment, based on patient

preference for cream rather than ointment formulations.

4.2. CREAM FORMULATIONS

4.2.1. Overview

Pharmaceutical emulsions have been described as systems that contain at least two immiscible

liquids, such as an oil and water and in which one of the phases is dispersed as globules viz., the

internal or dispersed phase, within the other phase viz., the external or continuous phase [168,

171, 172]. Since emulsions are formed from two incompatible liquids, they are considered to be

thermodynamically unstable colloidal systems and therefore usually require the addition of a

118

third component such as an emulsifying agent or surfactant to impart stability to the system [171,

172].

Emulsions are thermodynamically unstable since the two immiscible phases that exist in close

conjunction impart a positive interfacial free energy to the emulsion system [171-173]. As a

consequence of the positive interfacial free energy, the two immiscible liquids will always tend

to divide into their separate components in an attempt to achieve thermodynamic equilibrium

[173]. Therefore by use of an emulsifying agent, that is a compound possessing both polar and

non-polar characteristics that aligns itself at the interface between the oil and water phases, the

interfacial free energy between the phases can be reduced [171, 174]. The film structures formed

by an emulsifying agent at the interface between phases forms the basis of homogenous and

stable emulsions [171, 174].

Emulsions may be classed as either oil-in-water (o/w) or water-in-oil (w/o) systems [168, 171-

173] and a schematic representation of such emulsions is depicted in Figure 4.1. Other types of

emulsions that have been reported are the so-called multiple emulsions such as, for example,

water-in-oil-in-water (w/o/w) emulsions [175-178] and oil-in-water-in-oil (o/w/o) emulsions

[177-180]. Essentially, o/w emulsions contain oil droplets dispersed in an aqueous continuous

phase, whereas w/o emulsions contain aqueous or polar droplets dispersed in an oily continuous

phase [168, 171-173]. Molecules of an emulsifying agent assemble at the oil/water interface and

the hydrophilic or polar group of the emulsifying agent is orientated towards the water phase and

the hydrophobic or non-polar tail aligns itself towards the oil phase [171] as depicted in Figure

4.1.

Multiple emulsions such as w/o/w emulsions are emulsions in which a w/o emulsion is dispersed

as droplets in an aqueous phase [175, 176, 178], whereas o/w/o emulsions occur as emulsions in

which an o/w emulsion is dispersed in an oil phase [178-180]. In order to stabilize multiple

emulsions, two surfactants are invariably used, with a hydrophobic surfactant designed to

stabilize the interface of the w/o internal emulsion and a hydrophilic surfactant to stabilize the

external interface of the oil globules for w/o/w multiple emulsions, and the converse is true for

o/w/o multiple emulsions [178].

119

Oil

Water

Water

Oil

Oil inWater(O/W)

Waterin Oil(W/O)

Figure 4.1. Schematic representation of the principles of oil-in-water (o/w) and water-in-oil (w/o) emulsions (redrawn from 181).

The most common type of emulsions used in dermatological therapy are creams [171]. Creams

are described as emulsions of a semi-solid consistency [172, 182] or emulsions of a high

apparent viscosity [172, 182] with a typical creamy white appearance [183] that are

manufactured for topical application [172, 182]. However, more recently Buhse et al. [168]

classified and reported creams as semisolid dosage forms that contain >20% water and volatile

ingredients and/or <50% of hydrocarbon, wax or polyethylene glycol constituents as a vehicle

and that are intended for external application to the skin.

Buhse et al., [168] also described creams as semi-solid systems that display plastic flow

behaviour and in total contrast to other semi-solid formulations such as ointment bases, creams

exhibit two or more transition states when evaluated using thermogravimetric analysis (TGA).

The transition states are indicative of a system composed of at least two phases and cream

formulations can be classified as semi-solid systems that appear opaque, are viscous and have a

120

non-greasy to mildly greasy texture and tend to evaporate or get absorbed when rubbed onto the

skin [168].

Similar to liquid emulsions, creams usually contain a third component viz., an emulsion-

stabilizing system or an emulsifier [171]. However, in contrast to liquid emulsions, creams are

reported to contain more emulsifier than that required to form a condensed monomolecular

surfactant film at a droplet interface in liquid emulsions [172]. It has been argued that excess

emulsifiers in cream formulations interact with other components of these formulations either at

the droplet interface or in the bulk phase to produce complex, multiphase structures and these

complex multiphase structures are reported to be essential for the formation of creams that are

stable for extended periods of time [172] (Section 4.2.3.3., vide infra).

Pharmaceutical creams may contain one or more APIs dissolved or dispersed in either an o/w or

a w/o system [167]. Oil-in-water creams are usually referred to as “vanishing creams” as when

rubbed into the skin the formulation disappears without leaving any trace of their presence on the

skin [172].

4.2.2. Instability mechanisms in creams

4.2.2.1. Overview

The physical instability of creams has been reported to occur through various time- and

temperature-dependent physicochemical destabilizing mechanisms [171, 184-189] and these

mechanisms are summarized schematically in Figure 4.2.

121

flocc

ulat

ion

ostwald

ripeningph

ase

inve

rsio

n

creaming

sedimentation

coalescence

Figure 4.2. Schematic representation of the mechanisms by which creams show instability (adapted from 184)

4.2.2.2. Flocculation

In order for a cream to form, two immiscible liquid phases must be mechanically agitated in the

presence of an emulgent [171]. It has been reported that when agitation occurs in the absence of

any form of interfacial stabilization both phases of the cream will form droplets that will rapidly

flocculate and separate into two distinct phases [171, 184]. Flocculation may therefore be

described as the close accumulation of two or more individual droplets of a dispersed phase to

form loose assemblies or flocs, without loss of the interfacial film [171, 184]. In other words, in

flocculated cream systems the single droplets of the dispersed phase become replaced by twin

droplets or multiple flocs separated by a thin interfacial film [171, 184].

It has been reported that flocculation occurs as a result of van der Waal’s attractive forces that

take place in the absence of adequate repulsion between the droplets of the dispersed phase [171,

184]. Usually, individual dispersed phase droplets move through the external phase due to

122

diffusion or agitation [184]. If resistance between the droplets is not sufficient, flocculation

occurs and for example individual droplets may aggregate to form flocs [184] Flocculation does

not lead to an increase in the average size of the emulsion droplets, indicating that individual

droplets do not lose their integrity [186, 190].

4.2.2.3. Coalescence

The term coalescence is used to describe the aggregation of flocculated droplets into one large

droplet [171]. It has been argued that thinning and/or disruption or loss of the interfacial film

between approaching droplets in a cream layer is the main driving force for coalescence [171,

184]. Tadros [184] reported that when two flocculated droplets are in close proximity to each

other, the liquid surfaces undergo fluctuations forming what Tadros referred to as “waves”.

The apices of these fluctuations become the point at which strong van der Waal’s forces of

attraction are prevalent and when these fluctuations grow in amplitude the distance separating

the apices of the interfacial film may reach a critical value that will cause the film to collapse

[184]. Subsequently the two flocculated droplets combine to form a single larger droplet [184].

Unlike flocculation coalescence invariably leads to an increase in the average size of the droplets

[186, 190] as seen in Figure 4.2.

4.2.2.4. Creaming or sedimentation

Creaming refers to the process by which buoyant droplets of a dispersed phase rise to the top of a

container [184, 188, 189], and sedimentation takes place when dispersed droplets sink to the

bottom of a container [184]. Creaming or sedimentation occurs due to differences in the densities

of the dispersed and continuous phases [184, 188]. The dispersed phase is usually less dense than

the continuous phase and therefore droplets will cream or sediment in a gravitational field [184,

188]. Creaming or sedimentation does not involve an increase in the average size of the

dispersed droplets, although both creaming and sedimentation may occur prior to coalescence,

since coalescence requires the droplets of the dispersed phase to be in close proximity [186].

123

4.2.2.5. Ostwald ripening

Another relatively important but often neglected cream or emulsion instability is known as

Ostwald ripening [184-186, 190]. Ostwald ripening has been described as the growth of large

emulsion droplets at the expense of smaller ones due to differences in the solubility [184] or the

chemical potential [185, 186, 190] of small and large droplets.

The difference in chemical potential arises from the difference in the radius of curvature of the

droplets [184-186, 190]. Ostwald ripening may also be described as a mass transfer between

drops of different curvatures through a surrounding continuous phase [185]. The chemical

potential of a droplet is reported to increase with a decreasing radius of curvature and, as a

consequence, the solubility of the dispersed material at the surface of the droplet also increases

[185, 190].

It follows that materials at the surface of smaller droplets tend to dissolve and subsequently

diffuse through the continuous phase down a concentration gradient and are deposited on the

surface of larger droplets [185, 186, 190]. As a result of mass transfer from small droplets to

large droplets the small droplets shrink and ultimately disappear, whereas large droplets grow

eventually leading to the formation of a cracked or separated cream [185, 186, 190].

Ostwald ripening does not require the droplets to be in close proximity, since the process takes

place by transport of dissolved matter from one droplet to another, through the external phase

[186]. Ostwald ripening generally proceeds with the cube of the average radius of a droplet

varying linearly with time and this is one of the reasons why Ostwald ripening is usually not

considered as being an important phenomenon when considering macroemulsions, since droplets

in macroemulsions have radii in excess of between 1-2 µm [190]. Nevertheless, Ostwald ripening

is an important cause of instability in creams.

124

4.2.2.6. Phase inversion

Phase inversion is a phenomenon that occurs when one of the two phases of a cream, for

example the internal phase, becomes the external phase as the droplets of the internal phase

coalesce faster than the droplets of the external phase [171]. Phase inversion may also be caused

by an increase in the volume fraction of a dispersed phase or by a transition produced by

changing temperature and/or the addition of an electrolyte to a previously stable cream

formulation [184].

4.2.3. Stabilization of creams

4.2.3.1. Surfactants

The physical instability of creams invariably occurs due to the tendency of formulations to revert

back to the original distinct two phase systems that have a minimum interfacial free energy

(Section 4.2.2) [186]. Since interfacial free energy is the driving force for the irreversible fusion

of droplets as seen with flocculation and coalescence (Section 4.2.2), creams may be stabilized

by the inclusion of an appropriate emulsion-stabilizing system that will concentrate at the oil-

water interface [172, 173].. The use of emulsion stabilization results in a lowering of the

interfacial tension between oil and water phases [171, 182]. The stabilizing systems used in most

creams consist of surfactants or surface active agents (SAA) or amphiphiles [191].

SAA tend to settle at the boundary between two immiscible phases due to their chemical

structure [182]. SAA are characterized by having two distinct regions in their chemical structure

viz., a hydrophilic and hydrophobic region [182] as shown in Figure 4.1. The hydrophilic

portions may be ionic or non-ionic, whereas hydrophobic regions are invariably saturated or

unsaturated hydrocarbon chains or, less commonly, heterocyclic or aromatic ring structures

[182]. Generally, SAA are classified according to the nature of the hydrophilic group, therefore

SAA can be anionic, such as for example sodium alkyl sulphates, cationic, such as for example

alkylammonium halides, or non-ionic, such as polyoxyethylene alkyl ethers or polysorbates [182,

191].

125

Non-ionic SAA are reported to be the most widely used group of surfactants in cosmetic and

pharmaceutical creams [182]. Non-ionic surfactants do not possess a charged group in their

hydrophilic polar group, and therefore have a greater degree of compatibility with other

components of cream formulations than do the ionic surfactants [182]. In addition non-ionic

surfactants are reportedly less sensitive to changes in pH or to the addition of electrolytes, and

when used in topical cream formulations non-ionic surfactants tend to cause less skin irritation

than ionic surfactants [182]. However, it has been suggested that the main disadvantage of using

non-ionic surfactants is that they tend to be more expensive than ionic surfactants [182].

When incorporated into a cream formulation, the hydrophilic region of an SAA is orientated

towards the aqueous phase and the hydrophobic portion towards the oil phase [171] (Figure 4.1).

The resultant cream that is formed is either an o/w or a w/o emulsion and the specific

configuration that is formed depends on the properties of the emulgent system used to stabilize

the interface between the dispersed droplets and the continuous phase [171, 191]. The

emulsification capacity of SAA is reported to be determined by the relative difference in the size

and strength of the polar and non-polar groups that make up the molecule [182].

Oil in water creams are prepared if the hydrophilic characteristics of a SAA are slightly more

dominant than the hydrophobic characteristics since the SAA molecule will orientate at the

interface in such a way that the hydrophobic portion is forced to the centre of the unit [171, 182]

(Figure 4.1). Similarly, it has been reported that w/o creams are prepared if the hydrophobic

properties of an amphiphile are slightly more dominant than the hydrophilic characteristics of

that molecule [171, 182] (Figure 4.1).

4.2.3.2. Mixed emulgents

In the preparation of simple liquid emulsions, it has been suggested that surfactants alone can

stabilize such formulations and mixtures of surfactants have the ability to form more stable

emulsions than individual surfactants [171]. However the manufacture of a consistent cream

product with a realistic shelf-life may only be achieved by incorporating a specific mixed

emulsifying system into the cream formulation [173]. By definition, a mixed emulsifier is an

126

emulsifying system that consists of a combination of an ionic or non-ionic SAA and a fatty

amphiphile, such as a fatty alcohol, fatty acid or monoglyceride [173].

The combination of a surfactant with a fatty amphiphile in the correct ratio is able to produce a

powerful emulsifying system of the o/w type with excellent stabilization and thickening

properties [172]. It has been suggested that nine (9) parts by weight of fatty alcohol to one (1)

part by weight of an ionic surfactant (12:1 molar ratio) or four (4) parts by weight of a fatty

alcohol to one (1) part by weight of a non-ionic surfactant (20:1 molar ratio) may also be used to

prepare appropriate emulsifying blends for use in o/w creams [173]. The components of mixed

emulsifying systems may be added to a cream formulation separately during the manufacturing

process or alternatively as a previously blended mixture of emulsifying wax [172, 173].

In its simplest form a cream consists of an oil phase, a water phase and a mixed emulsifying

system [171, 192]. Commercial pharmaceutical o/w creams, however, are complex polydispersed

systems usually manufactured with several emulgents which complement the properties of each

other [172, 173]. Mixed emulgents that are used in o/w creams are usually water-soluble and

may consist of anionic or cationic SAA, such as sodium lauryl sulphate or cetrimide respectively,

and/or a non-ionic SAA, such as cetomacrogol 1000 in combination with fatty amphiphiles

[173].

The amphiphiles are usually higher fatty alcohols having chain lengths of between fourteen (14)

and eighteen (18) carbon atoms (C14 to C18) and may include substances such as cetyl, stearyl

and cetostearyl alcohols, glycerol monosterate and stearic acid [172, 173]. Typical

pharmacopoeial and commercially available emulsifying waxes and their respective

compositions are summarized in Table 4.1 [172, 173].

127

Table 4.1. Pharmacopoeial and commercially available emulsifying waxes [172, 173]

Emulsifying wax Constituents Emulsifying wax BP Cetostearyl alcohol and sodium lauryl sulphate Emulsifying wax USNF Cetostearyl alcohol and polysorbate-60 Cationic emulsifying wax BPC Cetostearyl alcohol and cetrimide Cetomacrogol emulsifying wax BPC Cetostearyl alcohol and cetomacrogol-1000 Glyceryl monostearate S.E. Glycerol monostearate and sodium citrate Lecithin Phosphatidylcholine, phosphatidylethanolamine,

phosphatidylinositol and phosphatic acid

4.2.3.3. Theory of cream emulsification

It has been argued that a reasonable and coherent explanation for the manner in which mixed

emulsifiers or emulsifying waxes stabilize oil droplets and control the consistency of o/w creams

may be given using the gel network theory of emulsion stability [171-173]. The gel network

theory relates the consistencies and stabilities of o/w creams to the presence or absence of

crystalline viscoelastic gel networks in the external phase of cream formulations [171-173].

The amount of mixed emulgent added to an o/w semi-solid pharmaceutical cream is usually in

excess of that required to form a monomolecular interfacial film in a simple model emulsion

[171, 173]. The surplus emulsifying wax in a cream formulation interacts with the aqueous

continuous phase to form strong crystalline viscoelastic gel networks at the oil-water interface of

the dispersed droplet [171, 173]. In general, the gel network theory illustrated schematically in

Figure 4.3 implies that a cream formulated to contain a mixed emulsifier system consisting of

combination of a fatty amphiphile, such as cetostearyl alcohol, and an ionic or a non-ionic

surfactant may be composed of at least four (4) phases [171, 173] viz.:

a) a bulk water phase,

b) a dispersed oil phase,

c) a crystalline hydrate phase,

d) a crystalline gel phase.

128

b. Surface of oil droplet

Interlamellar waterMultiple lamellae

a. Bulk water

d. Gel phase

c. Crystalline hydrateFatty amphiphile (eg, cetostearyl alcohol)

Surfactant (ionic or nonionic) Figure 4.3. Illustration of the gel network theory (adapted from 171).

Eccleston [173] reported that fatty amphiphiles such as cetostearyl alcohol usually exhibit

marked crystalline polymorphism and have the capacity to form an α-crystalline polymorph

under specific conditions. Eccleston [173] argued that in the α-form, the hydrocarbon chains of

the amphiphile are hexagonally packed and there is a rotation about the long axis of the

molecule. Eccleston [173] further argued that in the presence of water (a) the α-crystalline

polymorph will form hydrated crystalline phase (b) in which the thickness of the water layers

incorporated in the polar groups of the amphiphile is approximately eighteen (18) Å and is

limited by the considerable strength of the van der Waal’s attractive forces balancing osmotic

repulsion in the system.

The presence of very small quantities of ionic or non-ionic surfactant affect the degree of

swelling in the system and the swelling is enhanced giving rise to a swollen crystalline gel phase

(d) [172].. The swelling capacity of the crystalline gel phase depends on the hydrophilic

properties of the intrinsic polar groups of the amphiphile used in the emulsifier system [171]. It

129

has been suggested that the gel phase is characterized by a lamellar structure of alternating

bilayers of fatty amphiphile and surfactant separated by layers of interlamellar and fixed water

[171, 173].

The swelling properties and the concentration of the crystalline gel phase are factors that dictate

the overall consistency of a cream structure [171, 172]. The interlamellar-fixed water located

between the bilayers of the gel and crystalline hydrate phases enhances the volume ratio of the

dispersed phase to the free bulk water phase, resulting in stiffening of the system [173]. In order

to maintain the stability of a system it is vital to maintain a dynamic equilibrium between the

fixed water in the interlamellar gel and the bulk water phases [171, 172].

At low mixed emulsifier concentrations, such as for example 2-4% w/w, stiffening is sufficient

to produce a structured lotion [173]. However, at high mixed emulsifier concentrations, such as

for example > 4% w/w, the crystalline hydrate and the gel phases link to form strong a crystalline

viscoelastic gel phase and the system becomes a structured semi-solid cream [173].

The swelling capacity of the α-crystalline polymorphic form of an amphiphile that leads to the

formation of the crystalline viscoelastic gel phase is greater in the presence of ionic surfactants

than in the presence of non-ionic surfactants [171]. This has been ascribed to that fact that

swelling in the presence of an ionic surfactant is due to an electrostatic phenomenon, whereas in

the presence of a non-ionic surfactant, swelling is attributed to hydration of the polyoxythylene

chains of the surfactant and is usually limited by the length of the chain [171]. Nevertheless, the

underlying principles of the gel network theory apply whichever surfactant and fatty amphiphile

is used, albeit to different extents [173].

The gel network theory reveals that mixed emulsifiers through swelling of the α-crystalline form

of an amphiphile and subsequent formation of crystalline viscoelastic gel networks introduce an

electrostatic repulsive force between dispersed droplets [173]. Consequently, there is a reduction

in van der Waal’s forces of attraction between dispersed oil droplets which subsequently

prevents or delays the processes of flocculation, coalescence and/or Ostwald ripening (Section

4.2.2) from occurring [173]. Furthermore, the resultant stiffening of a cream structure due to the

130

formation of a crystalline viscoelatic gel phase invariably inhibits or reduces creaming or

sedimentation (Section 4.2.2) in a cream formulation [184].

The formation of a crystalline viscoelastic gel network phase in a cream formulation is

dependent on the type of amphiphile and the nature of the ionic or non-ionic surfactant, the

concentration of the mixed emulsifier and the molar ratios of amphiphile to surfactant used in

that formulation [173].

4.2.3.4. Hydrophilic-lipophilic balance (HLB)

Sections 4.2.3.2 and 4.2.3.3 report that the physical stability of pharmaceutical creams may be

best achieved using mixed emulsifiers rather than by use of a single SAA. In addition the

components of a mixed emulsifier may be added separately to a cream formulation during

manufacture, or alternatively as a previously blended emulsifying wax (Table 4.1). When the

components of a mixed emulsifier system are added separately, it is vital that the formulation

scientist determines the appropriate ratio of each component necessary to produce a physically

stable cream [182].

The relative quantities of individual emulgents may be calculated using a tool referred to as the

hydrophilic-lipophilic balance (HLB) [171, 182]. Essentially each emulgent is allocated an HLB

number representing the relative balance between the hydrophilic and lipophilic characteristics

within a SAA molecule [171, 182]. The HLB number is an arbitrary value ranging between 0-20,

which is then assigned to a particular SAA [171].

According to the HLB system, hydrophilic surfactants have high HLB values, whereas

hydrophobic surfactants have low HLB values [171, 182]. Generally emulgents with HLB values

ranging between 4-6 are w/o emulsifiers and the use of these emulsifiers tends to favour the

formation of w/o creams [171, 182]. SAA with HLB values ranging between 8-18 are reported

to be o/w emulsifiers and their use will favour the formation of o/w creams [171, 182]. Although

the HLB system was originally applied to non-ionic SAA its successful use has been extended

and applied to ionic emulgents [182].

131

The HLB number of a polyoxyethylene-based non-ionic SAA may be calculated using Equation

4.1 [171]:

5% cgrouphydrophobimolHLB = Equation 4.1.

The HLB value of a polyhydric alcohol fatty acid ester, such as glyceryl monostearate may be

derived from Equation 4.2 [171]:

⎟⎠⎞

⎜⎝⎛ −

=A

SHLB 120 Equation 4.2.

Where,

S = the saponification value of the ester

A = the acid value of the fatty acid

When it is impossible to obtain a saponification value of a compound as is the case with lanolin

derivatives, the HLB may be calculated using Equation 4.3 [171]:

⎟⎠⎞

⎜⎝⎛ +

=5

PEHLB Equation 4.3.

Where,

E = the weight percent (wt %) of the polyoxyethylene chain

P = the wt % of the polyhydric alcohol group

When using a mixed emulsifier system, the overall HLB value of a surfactant mixture (HLBM)

may be calculated using Equation 4.4 [171]:

( ) BAM HLBffHLBHLB −+= 1 Equation 4.4.

Where,

f = the weight fraction of emulsifier A.

132

The HLB system and theory may be used as an approximation in the design of a cream

formulation, however the stability of a cream may not be guaranteed by the use of a mixed

emulsifier system with an appropriate HLB value [171]. This is more than likely due to the fact

that a typical destabilization process, such as for example creaming, may be much more

dependent on other variables, such as viscosity of the continuous phase, rather than the

characteristics of the interfacial film [171]. Thus such factors should be taken into consideration

when designing complex pharmaceutical cream formulations.

4.3. EXPERIMENTAL

4.3.1. Characterization of CP creams

4.3.1.1. Overview

Prior to the development of a generic CP cream formulation, it was considered essential to select

physicochemical quality control tests for use in conjunction with an in vitro release test method

(Chapter 3) to characterize the commercial and extemporaneously manufactured CP cream

formulations. In addition the assessment of quality and consistency of extemporaneously

manufactured CP topical formulations form an integral part of formulation development studies.

Physicochemical tests such as the determination of solubility, particle size, size distribution and

crystalline form of an API and evaluation of the intrinsic viscosity and homogeneity of cream

products, have been traditionally used to provide reasonable evidence of consistent product

performance (Chapter 3). However, for the purposes of these studies, innovator and generic

cream products were characterized for CP content, apparent intrinsic viscosity, pH and in vitro

release.

4.3.1.2. Assay of CP content

The United States Pharmacopoeia (USP) [23] specifies that CP cream formulations should

contain not less than 90.0% and not more than 115.0% of the labelled amount of CP. As a

consequence, it was considered essential to assay CP content in all generic formulations

133

manufactured extemporaneously in these studies in order to ensure that the formulations

manufactured contained CP within this specified range. The CP content of all extemporaneously

manufactured and innovator cream products was assayed using the sample preparation procedure

described in Section 2.4.6.2 (Chapter 2) and as illustrated in Figure 2.7 (Chapter 2).

4.3.1.3. Intrinsic viscosity

The intrinsic viscosity of a semi-solid formulation invariably affects the rate and extent of release

of an active pharmaceutical ingredient (API) from a vehicle (Section 3.3.2.3). Therefore, the

evaluation of the intrinsic viscosity values of extemporaneously manufactured generic CP cream

products and the intrinsic viscosity value of the innovator product was considered essential in

these studies.

Any similarity or difference in the intrinsic viscosity values between the innovator and

extemporaneously manufactured CP cream products may be used as a basis for the discussion of

similarity or differences in the in vitro release rates of CP from the generic and innovator cream

products.

The intrinsic viscosity values of all extemporaneously manufactured CP products and the

intrinsic viscosity value of the innovator formulation were measured as described in Section

3.3.2.3.2.

4.3.1.4. pH-determination

The pH of dermatological semi-solid vehicles has been reported to affect the extent of

dissociation of ionisable API molecules and therefore the thermodynamic activity of the API,

partitioning and skin penetration [193]. CP is a non-ionisable molecule and does not dissociate

(Section 1.3.2). Therefore, it is highly unlikely that the pH of a CP topical formulation would

affect the release rate of CP from the vehicle or penetration and partitioning of CP into the skin.

134

Healthy human skin reportedly has a surface pH that ranges between 4-6 [193, 194] and a pH

gradient exists within the skin [193]. Although the pH of a topical formulation of CP may not

have any influence on the performance characteristics of the product, it was nevertheless

considered vital to develop a drug delivery system that would be pH-sensitive to the human skin.

Consequently the pH of the innovator and extemporaneously manufactured cream formulations

was determined to ensure that the manufactured products had an appropriate pH.

The pH of extemporaneously manufactured CP cream formulations and the innovator product

was measured using a Model GLP 21 Crison pH meter (Crison Instruments, Barcelona, Spain).

The pH was evaluated using 100 g of the cream and the measurements were taken within twenty-

four (24) hours of manufacture of the extemporaneous formulations. The pH measurements were

recorded in triplicate to generate an average pH value for each formulation.

4.3.1.5. In vitro release rate

The role of in vitro release rate testing during the development of topical semi-solid formulations

has been discussed in Chapter 3. In this chapter the equipment and conditions suitable for

assessing the in vitro release rate of CP from cream formulations were developed and validated.

The in vitro release test was essentially used as a marker for the successful development of the

generic cream formulation relative to the commercially available innovator product (Dermovate®

cream). Furthermore, the in vitro release test was used to determine batch-to-batch consistency

as well as in the evaluation of cream formulations used in accelerated stability studies (Chapter

5).

The in vitro release studies were conducted using the in vitro release test conditions presented in

Table 3.5 (Section 3.2.12). In vitro release rate profiles of CP were plotted as cumulative amount

of CP released per unit area (Q) vs. the square root of time (t1/2) or Q vs. t1/2 (Section 3.3.1).

Similarities or differences in the in vitro release rates of CP from the test formulations (generic

products) vs. Dermovate® cream were determined theoretically using Q and flux values (Section

3.3.1) and statistically using a non-parametric test which was described and reported in Section

3.2.11.

135

4.3.2. Innovator product characterization

4.3.2.1. Overview

Kanfer et al. [195] suggest that during the development of a generic drug product, it may be

beneficial to critically evaluate the innovator product with respect to qualitative and/or

quantitative assessments. Thus, prior to the development of a generic CP cream formulation, the

in vitro performance characteristics of the innovator product were evaluated and the data

generated from these studies were used as a general guide for the successful development of a

generic formulation. The purpose of conducting these studies was to ensure that the in vitro

performance characteristics of the generic CP cream formulations produced during the

development phase were equivalent to those of the innovator product.

4.3.2.2. Qualitative composition

The formulator of a generic product may consider using the same or similar inactive excipients

as in the innovator product [151]. Consequently, as part of the characterization phase of

Dermovate® cream, it was considered essential to evaluate the qualitative composition of the

product, as this could give an indication of the inactive excipients that could be considered for

use in the formulation of a generic CP cream product. The qualitative composition of

Dermovate® cream [196] is listed in Appendix 2.

4.3.2.3. CP content

Dermovate® cream was assayed for CP content using the sample preparation procedure described

and reported in Section 2.4.6.2 and from these studies, Dermovate® cream was found to have a

CP content of 103.9 ± 2.797% (2.69% RSD) (n = 3). These data reveal that Dermovate® cream

complies with the USP specifications for clobetasol proprionate cream [23], which states that

formulations should contain not less than 90.0% and not more than 115.0% of the labelled

amount of CP. The small % RSD is indicative of the homogeneity of the product and the

precision of the analytical method.

136


The intrinsic viscosity of Dermovate® cream was measured as described in Section 3.3.2.3.2 and

it was found that Dermovate® cream has an apparent intrinsic viscosity value of 44.467 ± 1.026

KcP (2.31% RSD) (n = 3). The measurements were recorded at room temperature (22˚C).


The apparent pH of Dermovate® cream was determined at room temperature (22˚C) as described

in Section 4.3.1.4 and from these studies, it was found that Dermovate® cream has an apparent

pH value of 5.05 ± 0.0321 (0.00637% RSD) (n =3). These data reveal that the apparent pH of

Dermovate® cream lies within a pH range of between 4-6, which is considered to be the pH

range of a healthy human skin (Section 4.3.1.4).

4.3.2.6. In vitro release rate studies

The in vitro release rate of CP from Dermovate® cream was assessed using the in vitro release

test conditions described in Table 3.5 and the resultant in vitro release rate profile of CP from

Dermovate® cream plotted as Q vs t½ is shown in Figure 4.4. These data reveal that the

cumulative amount of CP released (Q) over the 72-hour in vitro release studies was 36.69 ±

5.720 µg/cm2, which was equivalent to a cumulative % release of 24.19 ± 3.750%. The average

in vitro release rate of CP from Dermovate® cream or flux (n = 6) was found to be 4.924

µg/cm2/hr1/2.

137

y = 4.9236x - 5.7233R2 = 0.9941

0

5

10

15

20

25

30

35

40

45

0 2 4 6 8 10

Time (h)0.5

Q (u

g/cm

2)

Figure 4.4. In vitro release rate profile for CP from Dermovate® cream (n = 6)

The in vitro release rate or flux of CP from Dermovate® cream of 4.924 µg/cm2/h1/2 appears to be

fairly slow. Moreover the extent of CP release of 36.69 ± 5.720 µg/cm2 or 24.19 ± 3.750% seems

to be low considering the fact that the experiments were conducted over a 72-hour time period.

The apparent slow in vitro release rate of CP from Dermovate® cream and the relatively small

cumulative amount of CP released over the 72-hour experiment may be attributed to the

seemingly high apparent intrinsic viscosity value of Dermovate® cream (Section 4.3.2.4).

CP release from Dermovate® cream formulation, however, appears to fit the Higuchi model

described and reported in Section 3.2.8.2. According to the Higuchi model, a plot of Q vs. t1/2

would be linear for a drug in suspension in a semi-solid matrix with an infinite dose application

provided that such release is diffusion-controlled by the matrix material that comprises the semi-

solid dosage form. Therefore, judging from the R2 value viz., R2 = 0.9941 shown in Figure 4.4,

CP release from Dermovate® cream may be diffusion-controlled.

138

4.3.3. Generic product development

4.3.3.1. Overview

In order to gain market approval, a generic product cannot be superior to or better than the

innovator product against which it is tested and the product must meet certain criteria [151] to be

considered bioequivalent and therefore therapeutically equivalent to the brand name product

[151]. During the development of generic drug product formulations, it is vital to ensure that the

developed formulation will have the same therapeutic efficacy, safety and performance

characteristics as the innovator product it is tested against [151].

Although the objective of the current research work was not to determine bioequivalence or

therapeutic equivalence of the extemporaneously manufactured generic CP cream formulation

relative to Dermovate® cream, it was vital to ensure that the generic formulation was at least

equivalent to the innovator cream product in terms of the in vitro tests conducted on these

products.

Essentially formulation development studies for the generic CP cream product were comprised

of three stages:

a) Preliminary studies for which the objective was to manufacture a formulation that

remained physically stable for at least twenty-four (24) hours after manufacture.

b) In vitro release studies for which the objective was to establish whether the

formulation tested had an in vitro release rate that was similar to that of the innovator

product.

c) Characterization of the generic formulation for that formulation that had an in vitro

release rate profile similar to that of the branded product.

In addition, the best generic CP formulation was then manufactured on a relatively larger scale to

produce a batch with a sufficient number of samples for accelerated stability studies. The

purpose of the latter studies was to determine whether the formulation development process was

139

successful in producing a stable generic CP cream product. The data generated from these

studies are described and reported in Chapter 5 vide infra.

4.3.3.2. Excipients

4.3.3.2.1. Overview

The excipients used in formulation development studies of generic CP cream products and their

sources are listed in Appendix 3. The physicochemical characteristics and functions of each

excipient used in these formulation development studies are described in detail in the following

Sections.

4.3.3.2.2. Clobetasol 17-propionate

The description, physicochemical characteristics, structural activity relationships, clinical

pharmacology and pharmacokinetic aspects of clobetasol 17-propionate (CP) have been reported

and described previously (Chapter 1).

4.3.3.2.3. Propylene glycol

Propylene glycol, chemically known as 1,2-propanediol, has been described as a clear,

colourless, viscous liquid with a sweet, slightly acrid taste resembling that of glycerin [19, 140,

197, 198]. Propylene glycol may be used as a preservative, humectant, plasticizer, solvent or

stabilizer in a variety of pharmaceutical formulations [19, 140]. However in these studies

propylene glycol was used mainly as a water-miscible co-solvent for CP. Propylene glycol has a

better ability than glycerine to dissolve drugs such as the corticosteroids [140, 198]. Propylene

glycol is generally regarded as a relatively non-toxic material and is considered to be minimally

irritant when used in topical formulations [19, 140]. Propylene glycol has also been reported to

be chemically stable when stored in well-closed containers at cool temperatures [197, 140].

140

4.3.3.2.4. Sodium citrate

Sodium citrate is the non-proprietary name for sodium citrate dihydrate [199] and occurs as

odourless, colourless, monoclinic crystals or as a white crystalline powder with a cooling saline

taste [19, 199]. The solubility of sodium citrate dihydrate has been reported to be 1 in 1.5 parts of

water at room temperature (20˚C) [19, 197, 199] and 1 in 0.6 parts of boiling water (100˚C) [19,

199]. Sodium citrate is generally regarded as a stable, non-toxic and non-irritant excipient that

can be used as a pH adjusting agent [199]. Sodium citrate was used as a pH adjusting agent to

adjust the pH of the CP cream formulations.

4.3.3.2.5. Citric acid

Citric acid is also known as citric acid monohydrate and occurs as colourless or translucent

crystals, or as a white crystalline efflorescent powder [19, 200]. Citric acid is odourless [19, 200]

and has a strong acidic taste and an orthorhombic crystal structure [200]. The solubility of citric

acid is 1 in less than 1 part of water at room temperature (20˚C) [19, 197, 200]. Citric acid was

used in conjunction with sodium citrate as a buffering agent [200].

4.3.3.2.6. Geleol®

Geleol® (Gattefossé SAS, Saint-Priest Cedex, France) is the proprietary name for glyceryl

monostearate. Glyceryl monostearate is a white or almost white to cream-coloured, wax-like

solid in the form of beads, flakes or powder [19, 201]. Glyceryl monostearate is waxy to the

touch and has a slight fatty odour and taste [19, 201]. Commercially available glyceryl

monostearate contains a mixture of variable proportions of glyceryl monostearate and glyceryl

monopalmitate [19, 201]. Glyceryl monstearate is insoluble in water but is soluble in mineral and

fixed oils [19, 201]. Glyceryl monostearate can be used as an emollient, emulsifying agent,

solubilising agent and/or stabilizing agent in topical pharmaceutical formulations, [201].

Glyceryl monostearate and other fatty acid monoesters are however reported to be inefficient

emulsifiers, but may be useful emollients that can be readily emulsified by common emulsifying

141

agents by the addition of other fatty materials to a formulation [19, 201]. When incorporated in

topical formulations, glyceryl monostearate is generally regarded as a non-toxic and non-irritant

material [201] and should be stored in a tightly closed container in a cool, dry place protected

from light [201]. Self-emulsifying grades of glyceryl monostearate are available and have been

reported to be incompatible with acidic substances [201]. Glyceryl monostearate has a melting

point of ≥ 55˚C [19, 201].

4.3.3.2.7. Cetostearyl alcohol

Cetostearyl alcohol occurs as a white or cream-coloured unctuous mass or as almost white flakes

or granules and has a faint characteristic sweet odour [19, 202]. When heated, cetostearyl alcohol

melts to a clear, colourless or pale yellow-coloured liquid that is free of suspended matter [19,

202]. Cetostearyl alcohol is practically insoluble in water and is soluble in oils [19, 197, 202].

Cetostearyl alcohol may be used as an emollient, emulsifying agent or as a viscosity-increasing

agent in semi-solid formulations [19, 202].

Cetostearyl alcohol has been used to increase the viscosity of topical pharmaceutical

formulations and imparts body to both w/o and o/w creams [19, 202]. Furthermore cetostearyl

alcohol stabilizes creams and can act as a co-emulsifier, thereby reducing the amount of

surfactant required to form a stable emulsion [202]. When used in combination with surfactants,

cetostearyl alcohol forms creams with very complex microstructures, including liquid crystals,

lamellar structures and gel phases (Section 4.2.3.3) [202].

Cetostearyl alcohol is stable when stored in a well-closed container in a cool, dry place [202].

Cetostearyl alcohol is incompatible with strong oxidizing agents and metal salts, but is generally

a non-toxic material [202] and has a melting range of between 49-56˚C [19, 202] or between 48-

55˚C [19].

142

4.3.3.2.8. White beeswax

White beeswax is a chemically bleached form of yellow beeswax [19, 197, 203] and occurs as

tasteless, white or yellow or slightly yellow-coloured sheets or fine granules with some

translucence [203]. White beeswax has an odour similar to that of yellow beeswax although the

odour is less intense [203]. White beeswax is practically insoluble in water, but soluble in fixed

and volatile oils and can be used as a stabilizing and stiffening agent in semi-solid formulations

[19, 197, 203]. White beeswax stabilizes w/o emulsions [203] and increases the consistency of

semi-solid formulations, such as creams and ointments [19, 197, 203].

White beeswax is a non-toxic and non-irritant material [203], however hypersensitivity which

has been ascribed to contaminants in the wax has been reported [19, 203]. White beeswax is

incompatible with strong oxidizing agents and is stable when stored in well-closed containers

protected from light [203]. White beeswax has a melting point range of between 61-65˚C [19,

197, 203] and when heated to above 150˚C esterification occurs with a consequent lowering of

the acid value and elevation of the melting point of the material [203].

4.3.3.2.9. Chlorocresol

Chlorocresol occurs as colourless or almost colourless dimorphous crystals or as a crystalline

powder with a characteristic phenolic odour [19, 197, 204]. Chlorocresol is soluble as 1 in 260

parts of water at room temperature (20˚C) [19, 197, 204] and 1 in 50 parts of boiling water [19,

204] and is soluble in fixed oils [19, 197, 204]. Chlorocresol is primarily used as a preservative

in topical formulations and has bactericidal activity against both Gram-positive and Gram-

negative organisms, spores, moulds and yeasts [19, 197, 204]. Chlorocresol is stable at room

temperature (22˚C) and should be stored in a well-closed container, protected from light, in a

cool, dry place [204]. Chlorocresol on contact with strong alkali agents produces heat and fumes

that ignite explosively [204]. The melting point range of chlorocresol is between 55.5-65˚C [204]

or between 63-67˚C [19]

143

4.3.3.2.10. Estol® 1474

Estol® 1474 (Uniqema (Pty) Ltd, Bryanston, Johannesburg, Gauteng SA) is the proprietary name

for glyceryl stearate. Estol® 1474 is a white powder at 25˚C [205] and is reported to be oil

soluble and has an HLB value of 2.0-3.0 [205, 206]. According to the manufacturers of Estol®

1474 the material may be used as an emulsifier for o/w or w/o creams and can modify the

consistency of such formulations thereby increasing the stability of cream products [205].

4.3.3.2.11. Ritapro® 200

Ritapro® 200 (Rita, Crystal Lake, IL, USA) is the proprietary name for an emulsifying wax

containing a mixture of stearyl alcohol and ceteareth-20 [207]. Ritapro® 200 is sold as a white

flaky or wax-like material and is odourless [207]. Ritapro® 200 may be used as an emulsifier for

o/w creams and has a melting range of between 55-63˚C [207].

4.3.3.2.12. Emulcire® 61 WL

Emulcire® 61 WL (Gattefossé SAS, Saint-Priest Cedex, France) is the proprietary name for a

self-emulsifying wax containing a mixture of cetyl alcohol, ceteth-20 and steareth-20 [208, 209].

Emulcire® 61 WL is sold as white pellets [209] and has a melting range of between 45.6-50.5˚C

[208]. Emulcire® 61 WL is considered to be an o/w emulsifier that may be used for emulsifying

difficult-to-formulate, unstable active ingredients and can yield creams with excellent heat

stability [208].

4.3.3.2.13. Gelot® 64

Gelot® 64 (Gattefossé SAS, Saint-Priest Cedex, France) is the proprietary name of a self-

emulsifying base containing a mixture of glyceryl monostearate and polyethylene glycol-75

stearate (PEG-75 stearate) [208]. The physical appearance of Gelot® 64 has not been reported but

visual observation of Gelot® 64 reveals that the material occurs as yellow or off-yellow pellets.

Gelot® 64 has a melting range of 55.5-62.5˚C [208] and is an o/w emulsifier that may be used for

144

emulsifying difficult-to-formulate, unstable active ingredients and can yield creams with

excellent heat stability [208].

4.3.4. Formulation composition

All formulation compositions developed and tested during the development of generic CP cream

are listed in Table 4.2. The qualitative and quantitative formulae were adapted from a published

formulation [210] which is listed in Appendix 4. The main modification made to the formulation

summarized in Appendix 4 was the use of different primary mixed emulsifiers.

145

Table 4.2. Percentage composition of generic CP cream formulations developed and assessed in these studies

Component # Excipient CP001 CP002 CP003 CP004 CP005 CP006 1 Clobetasol 17-propionate 0.05000 0.05000 0.05000 0.05000 0.0500 0.0500 2 Propylene glycol 44.50 44.50 44.50 44.50 44.50 44.50 3 Sodium citrate 0.05000 0.05000 0.05000 0.05000 0.0500 0.0500 4 Citric acid 0.050000 0.050000 0.050000 0.050000 0.0500 0.0500 5 Geleol® pastilles 5.000 5.000 5.000 5.000 5.000 5.000 6 Cetostearyl alcohol 4.000 4.000 4.000 4.000 4.000 4.000 7 White beeswax BP 0.6000 0.6000 0.6000 0.6000 0.6000 0.6000 8 Chlorocresol 0.07500 0.07500 0.07500 0.07500 0.07500 0.07500 9 Estol® 1474* 1.000 - - - - - 10 Ritapro® 200* - 1.000 - - - - 11 Emulcire® 61 WL* - - 1.000 - - - 12 Gelot® 64* - - - 1.000 1.000 1.000 13 Propylene glycol 7.000 7.000 7.000 7.000 7.000 7.000 14 Propylene glycol 2.675 2.675 2.675 2.675 2.675 2.675 15 Purified water 35.00 35.00 35.00 35.00 35.00 35.00

* Primary emulsifier

146

4.3.5. Manufacturing methods

4.3.5.1. Overview

Generic CP cream formulations were manufactured using a manufacturing process adapted from

a manufacturing method that has been published [210]. This method of manufacture was used for

the preparation of all CP cream formulations developed and tested in these studies and is

described in the following sections and illustrated in Figure 4.5. All items were weighed using a

Model AE-163 Mettler top-loading analytical balance (Mettler Instruments, Zurich,

Switzerland).

4.3.5.2. Aqueous phase

Purified water (item 15) was heated to 90˚C in a beaker using a Model RCH IKA-Combimag

hotplate magnetic stirrer (Jankel & Kunkel KG, Staufen, Germany). The temperature of the

heated purified water was brought down to and maintained at 60˚C using a Model NB-34980

Colora Ultra-Thermostat water bath (Colora, Lorch, Germany) that had been previously set at

60˚C. Sodium citrate and citric acid (items 3 and 4) were dissolved in item 15 at 60˚C and the

resultant solution was mixed with propylene glycol (item 2). The temperature of the aqueous

phase was maintained at 60˚C using the water bath.

4.3.5.3. Oil phase

Geleol® pastilles (item 5) and Estol® 1474 (item 9) or Ritapro® 200 (item 10) or Emulcire® 61

WL (item 11) or Gelot® 64 (item 12) were melted together with cetostearyl alcohol (item 6),

white beeswax BP (item 7) and chlorocresol (item 8) while stirring in a beaker previously heated

to and maintained at 75˚C using the IKA-combimag hotplate magnetic stirrer (Section 4.3.5.2).

Once melted, the temperature of the molten oil phase was allowed to cool to 60°C and was

maintained at that temperature using a Colora Ultra-Thermostat water bath (Section 4.3.5.2).

147

4.3.5.4. Dispersed phase

The aqueous phase (Section 4.3.5.2) was then transferred to the oil phase (Section 4.5.5.3)

maintained at 60˚C. The mixture was stirred manually with a glass stirring rod for ten (10)

minutes at that temperature and was then homogenized at approximately 15,000 rpm using a

Model 6-105 AF Virtis homogenizer (Virtis Co., Gardiner, N.Y. USA) for five (5) minutes. The

temperature of the mixture was allowed to cool to 50˚C while stirring manually with the

manufacturing beaker placed in a water bath at 20˚C.

4.3.5.5. Drug phase

Clobetasol (item 1) was mixed with propylene glycol (item 13) in a beaker and sonicated using a

Model 8845-30 ultrasonic bath (Cole-Parmer Instrument Co., Chicago, IL, USA) for

approximately twenty-five (25) minutes or until a clear solution was obtained. The solution was

heated to and maintained at 50˚C using a Colora water bath. Once the temperature of the solution

had reached 50˚C the solution was added to the dispersed phase (Section 4.3.5.4) whilst

maintaining the temperature at 50˚C. The beaker that had previously contained the drug phase

was rinsed with propylene glycol (item 14), which had been heated and maintained at 50˚C. The

rinse solution was added to the dispersed phase (Section 4.3.5.4) and the mixture was manually

mixed for a further ten (10) minutes and then homogenized at 15,000 rpm for an additional five

(5) minutes at 50˚C.

4.3.5.6. Cream formulation

The formulation prepared as described in Section 4.3.5.5 was cooled to 30˚C with continual

manual stirring with the manufacturing beaker placed in a water bath at 20˚C. After cooling, the

cream was passed through a Model-HO valve type of homogenizer (Erweka-Apparatebau,

G.m.b.H, Heusenstamm, Germany) in order to generate a smooth formulation of improved

consistency. The cream was then packaged into 100 g ointment jars and stored at room

temperature (22˚C) until required for further analysis.

148

Figure 4.5. Schematic illustration of the manufacturing method for the CP cream formulation

Heat item 15 to 90˚C

Melt items 5, 6, 7, 8 and 9 or 10 or 11 or 12 at 75˚C

Cool down item 15 to 60˚C

Add items 2, 3 and 4

Cool down to 60˚C

Mix manually for 10 min

Homogenize at 15,000 rpm for 5 min at 60˚C

Cool down to 50˚C while mixing manually

Dissolve item 1 in item 13 at 50˚C, and rinse with item 14

Homogenize at 15,000 rpm for 5 min at 50˚C

Pack the cream in cream jars

Cool down to 30˚C while mixing manually

Pass cream through a valve type of homogenizer

149

4.3.6. Preliminary studies

In the preliminary stages of these studies, the objective was to manufacture a CP cream

formulation extemporaneously that showed no visible signs of physical instability, such as, for

example, cracking, creaming, phase inversion and/or bleeding of the cream base from the

container. Physical instability was evaluated immediately after manufacture and then twenty-four

(24) hours after manufacture and storage at room temperature (22˚C). Initial formulation

development was undertaken on small batches of only 200 g and any formulation that showed

signs of physical instability immediately and/or after twenty-four (24) hours of storage at room

temperature (22˚C) was considered unsuitable and therefore not considered for further

investigation.

The initial prototype cream formulation, Batch CP001 containing Estol® 1474 as the primary

emulsifier, was extemporaneously manufactured. However, although Batch CP001 appeared

homogenous, smooth and physically stable immediately after manufacture, signs of instability,

such as phase separation or cracking and bleeding of the base from the container were observed

twenty-four (24) hours after storage at room temperature (22˚C). Consequently, Batch CP001

was not characterized further in these studies. A batch summary record for Batch CP001 is

reported in Appendix 4.

It is possible that Batch CP001 failed to remain physically stable over the twenty-four (24) hour

storage period because Estol® 1474 is not a suitable primary emulsifier for CP cream

formulation. Despite the manufacturers of Estol® 1474 [205] stating that the emulsifier may be

used for the preparation of w/o and o/w in emulsions, the compound may only be useful for the

manufacture of liquid emulsions and not semi-solid emulsions, such as creams. This

characteristic may be due to the fact that Estol® 1474 contains no surfactant but only a single

amphiphile, glyceryl stearate (Section 4.3.3.2.10).

Estol® 1474 therefore cannot be considered as a mixed emulsifier and considering the fact that no

surfactant was incorporated into Batch CP001 as a separate ingredient, Estol® 1474 was not able

to form a mixed emulsification system in situ. Consequently, viscoelastic gel networks at the oil-

150

water interface that are reported to provide and maintain the stability of o/w creams (Section

4.2.3.3) were not formed. The apparent stability of Batch CP001 observed immediately

following manufacture may possibly be attributed to the presence of amphiphiles in the

formulation that thickened the formulation immediately on production. However, on standing the

formulation was not sufficiently viscous to maintain the physical stability over a prolonged

period due to the fact that no surfactant was included in the formulation.

In order to manufacture a physically stable generic CP cream extemporaneously that would

remain stable over a longer period of time, it may be necessary to add an appropriate surfactant

to the formulation or substitute Estol® 1474 for a commercially available mixed emulsifier

system, which would be the ideal approach. Consequently, three (3) batches of cream, Batches

CP002, CP003 and CP004 were manufactured using the commercially available mixed

emulsifiers Ritapro® 200 (Section 4.3.3.2.11), Emulcire® 61 WL (Section 4.3.3.2.12) and Gelot®

64 (Section 4.3.3.2.13). The quantitative compositions of these three batches are shown in Table

4.7 and the batch summary records for these formulations are shown in Appendix 4.

Batches CP002, CP003 and CP004 also appeared to be homogenous, smooth and physically

stable immediately after manufacture as had been observed for Batch CP001. However, unlike

Batch CP001 all three batches in which mixed emulsifiers were incorporated showed no signs of

physical instability twenty-four (24) hours after manufacture and storage at room temperature

(22˚C). A possible explanation for the immediate prolonged stability of these formulations could

be that Ritapro® 200, Emulcire® 61 WL and Gelot® 64 are all emulsifying waxes, containing

mixtures of fatty amphiphiles and surfactants. Therefore Ritapro® 200, Emulcire® 61 WL and

Gelot® 64 provided consistency and stability to the creams due to their ability to form crystalline

viscoelastic gel networks in the external phase of the cream formulations.

151

4.5.7. In vitro release studies

4.5.7.1. Overview

Once physically stable creams had been manufactured using Ritapro® 200 (Batch CP002),

Emulcire® 61 WL (Batch CP003) and Gelot® 64 (Batch CP004), the effects of these

commercially available synthetic mixed primary emulsifiers on the in vitro release rate of CP

from the creams were investigated. The purpose of conducting these studies was to determine

which of the cream formulations produced an in vitro release rate profile for CP that is similar to

that of Dermovate® cream. The similarity assessment was based on a statistical evaluation using

a non-parametric test described and reported in Section 3.2.11. Based on the results of these in

vitro release studies the most suitable mixed primary emulsifier system was selected and used to

manufacture additional generic CP cream formulations.

4.5.7.2. Effects of Ritapro® 200

The in vitro release of CP from a prototype formulation, Batch CP002, was evaluated and the

results are summarised in Table 4.3 and compared with the in vitro release data obtained for

Dermovate® cream. The resultant in vitro release rate profile of CP from Batch CP002 and from

Dermovate® cream is depicted in Figure 4.6.

Table 4.3. Cumulative amount of CP released and the average in vitro release rate (flux) of CP from Batch CP002 and Dermovate® cream (n = 6)

Product Q (µg/cm2)

Cumulative % released

Flux (µg/cm2/hr1/2

Batch CP002 63.54 ± 7.050 42.09 ± 4.700 8.681 Dermovate® cream 42.50 ± 6.540 27.50 ± 4.660 5.888

In vitro release rate data summarised in Table 4.3 and depicted in Figure 4.6 reveal that the in

vitro release rate of CP from Batch CP002 may not be considered equivalent to the in vitro

release of CP from Dermovate® cream.

152

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )

Dermovate® cream Batch CP002

Figure 4.6. In vitro release rate profile for CP from Batch CP002 and Dermovate® cream

The lower limit (L.L.) and upper limit (U.L.) of a confidence interval (C.I.) calculated as

described in Section 3.2.11 (Chapter 3) with Batch CP002 (test) and Dermovate® cream

(reference) are summarised in Table 4.4. If the products are to be considered equivalent then the

resultant limits for the C.I. calculated from experimental data should lie within the 75%-133%

limits for the products to be considered equivalent [49].

Table 4.4. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I) calculated using Batch CP002 (test) and Dermovate® cream (reference)

Confidence intervals (C.I.) L.L. (%) U.L. (%) C.I. limits 75.00 133.00 Dermovate® cream vs. Batch CP002 123.85 166.36

The data summarised in Table 4.4 reveal that the L.L. and U.L. of the C.I. calculated for Batch

CP002 do not fall within the 75% -133% C.I. limits. Consequently, the in vitro release rate

profile of CP for Batch CP002 is not equivalent to that of CP release from Dermovate® cream.

153

Based on these findings, Ritapro® 200 was not considered to be an ideal mixed emulsifier for use

in manufacturing CP cream formulations.

4.5.7.3. Effects of Emulcire® 61 WL

The in vitro release rate of CP from a prototype formulation, Batch CP003, was assessed and the

results are summarised in Table 4.5 and compared with the in vitro release data obtained for




Product Q (µg/cm2)


Flux (µg/cm2/hr1/2)



vitro release rate of CP from Batch CP003 may not be considered equivalent to the in vitro

release of CP from Dermovate® cream.

154

0

10

20

30

40

50

60

70

80

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )



The lower limit (L.L.) and upper limit (U.L.) of a confidence interval (C.I.) calculated with

Batch CP003 (test) and Dermovate® cream (reference) are summarised in Table 4.6. If the

products are to be considered equivalent then the resultant limits for the C.I. calculated from

experimental data should lie within the 75%-133% limits for the products to be considered

equivalent [49]

Table 4.6. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I.) calculated for using Batch CP003 (test) and Dermovate® cream (reference)

Confidence intervals (C.I.) L.L.(%) U.L. (%) C.I. limits 75.00 133.00 Dermovate® cream vs. Batch CP003 122.68 153.74

The data summarised in Table 4.6 reveal that although the L.L. the C.I. calculated for Batch

CP003 does fall within the 75% -133% C.I. limits, the U.L. does not lie within these limits.

Consequently, the in vitro release rate profile of CP for Batch CP003 is not equivalent to that of

155

CP release from Dermovate® cream. Based on these findings, Emulcire® 61 WL was not

considered to be an ideal mixed emulsifier for use in manufacturing CP cream formulations.

4.5.7.4. Effects of Gelot® 64

The in vitro release rate of CP from a prototype formulation, Batch CP004, was evaluated and

the results are summarised in Table 4.7 and compared with the in vitro release data obtained for




Product Q (µg/cm2)


Flux (µg/cm2/hr1/2



vitro release rate of CP from Batch CP004 may be considered equivalent to the in vitro release of

CP from Dermovate® cream.

156

0

10

20

30

40

50

60

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )



The lower limit (L.L.) and upper limit (U.L.) of a confidence interval (C.I.) calculated with

Batch CP004 (test) and Dermovate® cream (reference) are summarised in Table 4.8. If the

products are to be considered equivalent then the resultant limits for the C.I. calculated from

experimental data should lie within the 75%-133% limits for the products to be considered

equivalent [49].

Table 4.8. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I) calculated using Dermovate® cream (reference) and Batch CP004 (test)

Confidence intervals (C.I.) L.L. (%) U.L. (%) C.I. limits 75.00 133.00 Dermovate® cream vs. Batch CP004 95.50 126.12

The data summarised in Table 4.8 reveal that the L.L. and U.L. of the C.I. calculated for Batch

CP004 do fall within the 75% -133% C.I. limits. Consequently, the in vitro release rate profile of

CP for Batch CP004 is equivalent to that of CP release from Dermovate® cream. Based on these

157

findings, Gelot® 64 was considered to be a suitable mixed emulgent for use in manufacturing CP

cream formulations.

4.3.8. Generic product characterization

4.3.8.1. Overview

Following the selection of the most suitable synthetic mixed emulsifier for use and the

development of a generic cream formulation containing CP which had the desirable in vitro

release rate profile, the next step was to characterize the physico-chemical properties of the

generic formulation and compare the data to those obtained for Dermovate® cream. Therefore

three (3) batches of 500 g viz., Batches CP004, CP005 and CP006, all containing equal amounts

of Gelot® 64 and manufactured using the manufacturing method described in Section 4.5.5, were

evaluated for CP content, viscosity, pH and in vitro release rate. The data generated from these

studies were also used to determine batch-to-batch uniformity and therefore the consistency of

the manufacturing method that had been developed. Batch summary records for Batches CP005

and CP006 are shown in Appendix 4.

4.3.8.2. CP content

The assay for CP content in Batches CP004, CP005 and CP006 was evaluated using the sample

preparation procedure described in Section 2.4.6.2 and the resultant data are listed in Table 4.9

and compared to the CP content of the innovator product Dermovate® cream.

Table 4.9. CP content of Dermovate® cream and Batches of generic cream formulations (n = 3)

Product CP content (%) SD % RSD Dermovate® cream 103.9 2.797 2.69 Batch CP004 109.8 1.114 1.06 Batch CP005 109.8 2.474 2.25 Batch CP006 113.1 2.605 2.30

158

The data reveal that the CP content in all the three generic batches manufactured comply with the

USP specifications [23] as does Dermovate® cream. Therefore the manufacturing process is

suitable for the production of the generic cream.


The intrinsic viscosity of Batches CP004, CP005 and CP006 was investigated using the

procedure described and reported in Section 3.3.2.3.2 (Chapter 3). The results generated from

these studies are summarised in Table 4.10 and compared to the intrinsic viscosity data of

Dermovate® cream. It appears that there is little difference in the intrinsic viscosity values among

the three batches manufactured, implying that the method of manufactured was consistent.

Table 4.10. Intrinsic viscosity readings for Dermovate® cream and generic cream products (n = 3)

Product Intrinsic viscosity (KcP) SD % RSD Dermovate® cream 44.47 1.026 2.31 Batch CP004 15.50 4.359 2.81 Batch CP005 12.37 0.3214 2.60 Batch CP006 11.43 0.05774 0.505

The data in shown in Table 4.10 clearly reveal a marked difference between the intrinsic

viscosity of Dermovate® cream and of any of the three extemporaneously manufactured generic

batches.

The intrinsic viscosity of Dermovate® cream is approximately three (3) times greater than the

intrinsic viscosity of Batches CP004, CP005 and CP006. The difference in the intrinsic viscosity

of the innovator formulation and of the three extemporaneously manufactured batches may be

attributed to either the gel network theory or to manufacturing process variables such as

homogenization speed and mode of cooling.

The gel network theory relates the consistency and stability of o/w creams to the presence or

absence of viscoelastic gel networks in the external phase of cream formulations. The crystalline

viscoelastic gel networks form in the continuous phase of a cream formulation when a fatty

159

amphiphile in a mixed emulsifier swells in the presence of water and in the presence of a

surfactant in a cream formulation. The presence of swollen crystalline gel networks and their

concentration in the continuous phase of a cream are reported to be responsible for the overall

consistency of a cream structure [171, 172] as well as the self-bodying action of the mixed

emulsifier [173].

Although an amphiphile will swell in the presence of any type of surfactant to form gel

networks, the amount of swelling of an amphiphile is reportedly greater in the presence of ionic

surfactants than in the presence of non-ionic surfactants [171]. In the presence of a non-ionic

surfactant, swelling has been reported to be attributed to the hydration of hydrophilic chains of

the non-ionic surfactant and is usually limited by the length of the chain [171].

The composition of the innovator formulation (Appendix 2) reveals that Arlacel® 165 is the

mixed emulsifier in Dermovate® cream. The mixed emulsifier in the extemporaneously

manufactured CP generic formulations is Gelot® 64. Both Arlacel® 165 [206] and Gelot® 64

(Section 4.3.3.2.13) contain a mixture of an amphiphile and a non-ionic surfactant. Essentially,

Arlacel® 165 is a mixture of glyceryl monostearate and polyethylene glycol-100 stearate or PEG-

100 stearate [206] whereas Gelot® 64 consists of glyceryl monostearate and polyethylene glycol-

75 stearate or PEG-75 stearate (Section 4.3.3.2.13).

It is clearly evident that both Arlacel® 165 and Gelot® 64 are comprised of similar ingredients,

the only difference being attributed to the chain length of the non-ionic surfactant. It is therefore

possible that the higher viscosity of Dermovate® cream compared to that of the generic CP

creams may be due to the increased length of the chain of the non-ionic surfactant in Arlacel®

165. The increased chain length more than likely resulted in greater swelling of the glyceryl

monostearate and subsequent formation of a large amount of the crystalline viscoelastic gel

phases in the continuous aqueous phase of the cream markedly increasing the viscosity of the

formulation.

The apparent high intrinsic viscosity of Dermovate® cream may also be attributed to

manufacturing process variables, such as the homogenization speed and mode of cooling. It has

160

been reported [170, 192] that the apparent intrinsic viscosity of a semi-solid formulation is

influenced by the size of the droplets of the dispersed phase and it follows that droplets of small

sizes will produce creams of higher viscosity than those of dispersed phases with large droplet

sizes [170, 192, 211].

A reduction in the size of dispersed droplets of a dispersed phase may be achieved by increasing

the speed of homogenization during manufacture [170, 192, 211]. The higher shearing forces

provided by increased homogenization speeds disrupts the hydrocarbon chains of the oil and wax

droplets, thereby exposing surfactant chains to water which in turn results in the formation of an

additional gel phase and subsequently further increases the intrinsic viscosity of a cream

formulation [170].

Although the speed of the laboratory homogenizer used in the manufacture of generic CP batches

(Section 3.3.5) was set to approximately 15,000 rpm it is likely that the laboratory homogenizer

does not provide the same shear as would be expected from high speed homogenizers used in the

large scale manufacture of cream products. Consequently, the apparent intrinsic viscosity values

of the extemporaneously manufactured CP batches did not match the apparent intrinsic viscosity

value of the commercially available Dermovate® cream.

As far as mode of cooling is concerned, Niellound et al, [212] established and reported that a

progressive mode of cooling of cream formulations at room temperature (25˚C), corresponding

to 45 minutes of homogenization time, appears to lead to the production of creams with higher

viscosity, better homogeneity and better stability than what Niellound et al, [212] referred to as

brutal mode of cooling with a water bath at 15˚C, corresponding to 30 minutes of

homogenization time.

It possible that at an industrial scale, a progressive mode of cooling corresponding to longer

homogenization time is adopted, resulting in cream formulations with higher viscosity as seen in

the case of Dermovate® cream. However, progressive cooling at a laboratory scale would be a

time-consuming and laborious process, especially in these studies where manual mixing while

cooling was used. Nevertheless, it is probable that the rapid cooling with a water bath at 20˚C as

161

was done in these studies and which corresponded to approximately 20 minutes of manual

mixing may have contributed to the formation of less viscous creams compared to Dermovate®

cream.


The pH of Batches CP004, CP005 and CP006 cream products was determined at room

temperature (22˚C) as described in Section 4.3.4. The data generated from these studies are listed

in Table 4.11 and compared to the apparent pH determined for Dermovate® cream. The pH

values for all three CP cream batches fall within the pH range of 4-6 of a healthy human skin and

therefore all the three formulations are unlikely to exert pH effects on pH-sensitive human skin.

The pH data of all three batches is consistent, implying that the method of manufacture was

consistent. In addition the pH of the creams is similar to that of Dermovate®.

Table 4.11. pH readings for Dermovate® cream and generic cream formulations (n = 3)

Product pH SD % RSD Dermovate® cream 5.05 0.0321 0.00637 Batch CP004 5.20 0.01527 0.294 Batch CP005 5.11 0.005770 0.113 Batch CP006 5.15 0.0264 0.514

4.3.8.5. In vitro release rate

The in vitro release of CP from Batches CP004, CP005 and CP006 was assessed and the

resultant data generated from these in vitro release studies are summarised in Table 4.12 and are

compared with the in vitro release data generated for Dermovate® cream. The resultant in vitro

release rate profiles of CP from Batches CP004, CP005 and CP006 and from Dermovate® cream

are depicted in Figure 4.9.

162

Table 4.12. Cumulative amount of CP released and the average in vitro release rate (flux) of CP from Dermovate® cream and generic cream formulations (n = 6)

Product Q (µg/cm2)


Flux (µg/cm2/hr1/2

Dermovate® cream 42.50 ± 6.540 27.50 ± 4.660 5.888 Batch CP004 49.46 ± 4.010 32.98 ± 2.640 6.515 Batch CP005 56.73 ± 2.82 37.83 ± 3.39 7.532 Batch CP006 43.99 ± 5.270 29.36 ± 3.740 5.796

0

10

20

30

40

50

60

70

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )

Dermovate® cream Batch CP006 Batch CP004 Batch CP005

Figure 4.9. In vitro release rate profile for CP release from CP generic cream products and Dermovate®

cream

From the in vitro release rate profile data for CP depicted in Figure 4.8 and summarised in Table

4.12, it appears that there is no significant difference in the in vitro rate of release of CP from

Batches CP004, CP005 and C006 and the release rate of CP from the innovator product,

Dermovate® cream. These data are consistent with the results of the non-parametric statistical

test (Table 4.13), which reveal that all the three manufactured batches are equivalent to

Dermovate® cream in terms of their CP in vitro release rate profiles.

163

Table 4.13. The lower limit (L.L.) and upper limit (U.L.) of the confidence interval (C.I.) calculated using Dermovate® cream (reference) and generic formulations (tests)

Confidence intervals (C.I.) L.L. (%) U.L. (%) C.I. limits 75.00 133.0 Dermovate® cream vs. Batch CP004 95.50 126.1 Dermovate® cream vs. Batch CP005 107.9 131.9 Dermovate® cream vs. Batch CP006 81.40 113.4

It is interesting to note that the in vitro release profile for CP from all three generic batches

appears to be similar to that of CP release from Dermovate® cream despite the apparent viscosity

of the innovator formulation being three times higher than the viscosity of any of the three

generic Batches (Table 4.10). From these observations, it may be implied that the intrinsic

viscosity of the cream formulation does not influence the in vitro release rate of CP in these

formulations significantly, which contradicts well-established theories on drug release from

semi-solid dosage forms [213].

It has been established that the rate and extent of release of an active pharmaceutical ingredient

(API) from a semi-solid vehicle decreases progressively as the intrinsic viscosity of the base

material increases, and the converse is also true [213]. Therefore it was expected that the

difference in the intrinsic viscosity of Dermovate® cream and the extemporaneously

manufactured CP cream batches would result in significantly different in vitro release rates for

CP from each of the products.

It has also been reported that for an API molecule to be released from a semi-solid vehicle, it

first has to reach the surface of the formulation and that mass transport in the cream formulation

may occur by diffusion of the API molecules or by diffusion and convection of oil droplets

[214].

The apparent intrinsic viscosity or macro-viscosity of a semi-solid product influences the

diffusion of oil droplets in the dosage form when the diffusion length is greater than the diffusion

length scale of the structure elements of a formulation [214]. In addition it has been suggested

that diffusion of API molecules through a semi-solid formulation is also influenced by the micro-

viscosity of a topical formulation and that the formulation environment may affect an API

164

molecule, such as for example the physical properties of the emulsifier used to manufacture a

stable dosage form [214].

It is possible that the in vitro release of CP from cream formulations may to a large extent be

affected by the physical properties of the mixed emulsifier system used in the formulation rather

than the apparent intrinsic viscosity of the cream formulation. The components of the mixed

emulsifier (Arlacel® 165) used in Dermovate® cream are essentially similar to the components of

the mixed emulsifier (Gelot® 64) used in the manufacture of the generic batches. Therefore the

behaviour of the mixed emulsifier in the cream formulation may have a dominant effect on the in

vitro release rate of CP from semi-solid cream formulations, rather than the macro-viscosity of

the semi-solid base.

4.4. CONCLUSIONS

The objective of these studies was to develop and assess a generic cream formulation for the

delivery of clobetasol 17-propionate (CP) for use in the treatment of patients with inflammatory

skin conditions, such as acute and chronic attacks of any type of eczema.

Initial experiments were designed to facilitate the characterization of the innovator product,

Dermovate® cream. Consequently, the innovator product was characterized in terms of

clobetasol 17-propionate (CP) content, apparent intrinsic viscosity, pH and in vitro release rate of

CP. The data generated from these studies were used as a benchmark for the development of

extemporaneously manufactured generic CP cream batches. The CP content of Dermovate®

cream was found to fall within the USP specifications, and the pH of the product was found to lie

in the pH range of the human skin.

Preliminary studies in the development of generic formulations were designed to facilitate the

manufacture of a prototype generic formulation that showed no signs of physical instability, such

as, for example, cracking, creaming, phase inversion and/or bleeding of the cream base from the

container, immediately after the manufacture as well as twenty-four (24) hours after manufacture

and storage at room temperature (22˚C).

165

Although a prototype formulation (Batch CP001) manufactured with Estol® 1474 as the primary

emulsifier showed evidence of physical stability immediately after manufacture, the formulation

was found to be physically unstable twenty-four (24) hours after manufacture and storage at

room temperature (22˚C) and was therefore not considered for further investigation.

Prototype formulations containing commercially available mixed emulsifiers, such as Ritapro®

200 (Batch CP002), Emulcire® 61 WL (Batch CP003) and Gelot® 64 (Batch CP004) were found

to be physically stable immediately after manufacture as well as twenty-four (24) hours after

manufacture and storage at room temperature (22˚C) and were therefore considered for further

evaluation. Once physically stable creams had been manufactured the in vitro release rate of CP

from Batches CP002, CP003 and CP004 and Dermovate® cream was evaluated.

The purpose of these studies was to identify a prototype batch with a similar in vitro release

profile for CP that was similar to that for Dermovate® cream. It was found that Batch CP004 was

the only batch with an in vitro release profile for CP that was similar to that of Dermovate®

cream. The similarity of release profiles was assessed using a non-parametric statistical test

recommended by the Food and Drug Administration (FDA).

The successful manufacture of Batch CP004 necessitated the production of two additional

batches of cream. Batches CP005 and CP006 were manufactured using Gelot® 64 as a mixed

emulsifier with a composition exactly the same as that used to manufacture Batch CP004.

Batches CP004, CP005 and CP006 were characterised in terms of CP content, apparent intrinsic

viscosity, pH and in vitro release rate testing. The CP content of all three extemporaneously

manufactured CP batches was within the specified USP limits and all the three batches had pH

values within the pH range of healthy human skin.

However, it was found that the apparent intrinsic viscosity of the innovator product was almost

three (3) times greater than that of all three generic cream batches. The difference in intrinsic

viscosity of Dermovate® cream was attributed to the physicochemical characteristics of the

mixed emulsifiers used in the innovator product and possibly to manufacturing process variables,

such as homogenization speed and mode of cooling.

166

Despite the difference in the apparent intrinsic viscosity of Dermovate® cream and the CP

batches, the in vitro release rate of CP was found to be statistically similar to the innovator

product for all three generic batches tested. It was concluded that the behaviour of the mixed

emulsifier in both the commercially available cream and the extemporaneously manufactured

creams, and not the apparent intrinsic viscosity of the formulations, has an impact on the in vitro

release rate of CP from these formulations.

Generic CP cream formulations have been successfully developed and characterized in vitro in

terms of CP content, apparent intrinsic viscosity, pH and in vitro release rate testing. Although

the viscosity of the generic CP cream formulations did not match the apparent viscosity of

Dermovate® cream, the generic CP batches were equivalent to the innovator cream product in

terms of other in vitro performance characteristics such as CP content, pH and in vitro release

rate.

The data generated from these studies formed the basis for manufacture of a larger batch of

cream that was subjected to accelerated stability studies at 40°C/25% RH (Chapter 5).

167

CHAPTER FIVE

STABILITY OF CLOBETASOL 17-PROPIONATE CREAMS

5.1. INTRODUCTION

Every medicinal product on the market is required to have an expiry date allocated by the

regulatory authorities in a particular country and the expiration date must be presented on the

label of the container in which the product is stored [215, 216]. The expiration date can be

defined as the time interval over which a pharmaceutical product remains within the

specifications established for product strength, quality and purity [215-217]. The expiration date

of a product may be determined by adding a product’s shelf-life period to the date of

manufacture [216, 217]. The shelf-life of a product specifically refers to the time period during

which the strength of the active pharmaceutical ingredient (API) in a product remains ≥ 90% of

the original label claim and is therefore the period for which a product will be suitable for use by

a patient [215, 217].

In order to determine the shelf-life of a medical product, pharmaceutical manufacturers must

conduct appropriate stability studies [216]. In addition to generating data on which to base

proposals for a product’s shelf-life, stability studies may also reveal information that is vital for

the selection of product packaging and storage conditions [217-219]. Stability studies may also

provide evidence on how the quality of a drug product varies with time under the influence of a

variety of different environmental factors, such as, for example, temperature, humidity and light

[219-221]. Evaluation of the stability of a pharmaceutical product is therefore considered an

integral part of formulation development studies [217, 218].

Most API molecules that are used in medicinal products are inherently unstable compounds

[218]. This can be ascribed to the presence of reactive functional groups, which, in addition to

providing reactive sites necessary to produce a therapeutic effect in vivo, may also increase the

susceptibility of a drug molecule to chemical reactions outside the body, leading to degradation

and subsequent loss of potency of a drug product [215]. Pharmaceutical products have been

168

reported to be susceptible to degradation via three (3) primary mechanisms: chemical, physical

and biological. These degradation mechanisms may occur in isolation or in combination [215,

222].

Chemical degradation in pharmaceutical systems invariably results in chemical changes to the

product. Such changes include for example the formation of degradation products, loss of API

potency and/or loss of activity of excipients such as anti-microbial preservatives and/or anti-

oxidants [219]. These changes may occur as a result of intrinsic chemical reactions, such as

hydrolysis, photolysis, oxidation, reduction and/or racemisation reactions [215, 222]. It has been

suggested that intrinsic chemical reactions may be triggered by extrinsic factors and therefore the

chemical stability of medicinal products may be regarded as the ability of a drug product to

withstand the effects of moisture, light, humidity and heat [219].

The physical degradation of pharmaceutical products may be caused by a range of external

factors, such as, impact, vibration, abrasion and/or temperature fluctuations [222]. Physical

degradation of a drug product may lead to changes in product appearance, consistency,

uniformity, dissolution rate, colour, odour, moisture content and/or pH amongst others and such

changes are usually dependent on the dosage form under investigation [219]. Biological and/or

microbiological degradation in medicinal products may lead to the proliferation of micro-

organisms in non-sterile products and changes in the efficacy of preservatives [219] and may be

caused by microbiological and/or non-microbiological organisms [222].

Whether a change in a delivery system is chemical, physical or microbiological such changes are

likely to adversely affect any attribute of the quality of the product in terms of its fitness for use

by a patient [215]. Therefore the stability of a pharmaceutical product may be defined as the

ability of a formulation in a specific container to remain within the physical, chemical,

microbiological and hence therapeutic, toxicological, protective and informational specifications

for that product [215, 217, 221]. In other words, the stability of a pharmaceutical product is the

extent to which a pharmaceutical product retains the same properties and characteristics that the

product possessed at the time of manufacture, throughout the shelf-life of the product within

specified limits [215, 217].

169

The factors that may adversely affect the stability of a pharmaceutical product include for

example the [215, 217]:

1) stability of the API,

2) potential interaction between an API and inactive excipients,

3) manufacturing process,

4) dosage form,

5) container/closure system,

6) environmental conditions the product is exposed to during shipment,

7) storage conditions,

8) handling, and,

9) length of time between manufacture and use.

During the formulation development process and manufacturing procedure a formulation

scientist may be able to control the external factors to which a dosage form or the components of

the dosage form may be exposed and that may affect product stability, and by control of these

parameters, minimize or prevent product instability [223]. However once a dosage form has left

a manufacturing plant, the formulation scientist has little or no control if any over the variables

that the dosage form in question may be exposed to during distribution, storage and use [223].

It may be possible to minimize the potential hazards to which the dosage form may be exposed

to during distribution, storage and use by selecting an appropriate package or packaging

configuration for the product [223]. However of all the environmental attributes that may be

involved in the degradation of an API in a dosage form, temperature is the most important factor

and cannot be controlled by the selection of specific packaging [215], and it has been argued that

the selection of a specific packaging configuration will not protect a pharmaceutical formulation

form from the detrimental and adverse effects associated with the use of elevated temperatures

[224]. The lack of protection from elevated temperatures afforded by packaging is problematic

for pharmaceutical products destined for used in high temperature zones, such as for example

Zones III or IV [218, 225].

170

In order to assess how a newly developed dosage form behaves under the influence of different

environmental conditions and to minimize or avoid unforeseen adverse effects it is essential to

assess the stability of the new pharmaceutical product in the desirable packaging configuration

that is intended for use during marketing of the product. Stability studies are usually conducted

under ambient and accelerated conditions [215-218, 221] and three different conditions for

stability studies have been reported [217, 219, 221]. However only two types of stability studies

viz., short-term and long-term have been widely reported [215-218, 221].

Long-term stability studies are usually conducted under ambient storage conditions [215, 216,

218, 221] and are defined as experiments for the evaluation of the physical, chemical, biological,

biopharmaceutical and microbiological characteristics of a drug substance or product, during and

beyond the expected shelf-life and the storage of samples at the recommended storage conditions

[217, 219]. Stability data generated from long-term stability studies are invariably the primary

stability data used to establish the shelf-life of a product and to confirm a projected shelf-life

and/or the recommended storage conditions for a product [216, 217, 219].

In contrast short-term or accelerated stability studies are experiments conducted under

exaggerated storage conditions, such as, for example, high temperature and/or relative humidities

[215-218, 221]. Accelerated stability studies are designed to increase the rate of a potential

chemical and/or physical change of an API or medicinal product such that significant

degradation, if it occurs, can be observed in a relatively short period of time [216, 219]. Data

generated from short-tem stability studies may be used in conjunction with long-term stability

studies to assess the longer term chemical effects at non-accelerated conditions and to evaluate

the impact of short term deviations outside the appropriate storage conditions of a product, such

as might occur during transportation, on product quality [217, 219, 226].

The objective of these studies was to conduct stability studies at elevated temperatures on the

clobetasol 17-propionate (CP) cream formulation developed and characterized as described and

reported in Chapter 4. Although laboratory-scale stability data are not normally acceptable as

primary stability data for use to establish product shelf-life and/or storage conditions, such data

may be submitted to regulatory agencies to support primary stability data [218, 219].

171

It is however important to note that the purpose of these stability studies was not to generate

primary stability data for use to estimate a shelf-life for the CP cream product but rather to

determine whether or not the formulation development process was successful in producing a

potentially stable CP cream product. The stability data generated for CP generic cream from

these studies would then form the basis for further more comprehensive studies on the stability

of the formulation.

5.2. EXPERIMENTAL

5.2.1. Overview

The initial step in the generation of stability data requires that a stability protocol or a detailed

stability testing plan is documented [218]. Since stability testing conditions may vary depending

on the inherent stability of a drug compound, type of the dosage form and/or the proposed

container-closure configuration to be used, the final stability protocol invariably relies on the

type of drug substance or drug product being tested [218].

The ultimate stability protocol will also depend on whether a medicinal product containing a

specific API is already on the market [218]. As a consequence some guidelines on stability

studies may require less stringent stability evaluations for drug substances and/or drug products

that have been on the market for extended period of time [218]. In addition, the target market for

a medicinal product must also be taken into account when establishing a stability protocol as

room temperature conditions vary from location to location and will affect the storage condition

requirements in addition to other test parameters [217, 218].

The World Health Organization (WHO) [217], International Conference on Harmonisation

(ICH) [221] and the Food and Drug Administration (FDA) [227] have published general

guidelines pertaining to the assessment of stability of drug substances and/or drug products. The

guidelines were used as a template to design a stability protocol for use in the assessment of

stability of the CP topical cream formulation developed in these studies.

172

5.2.2. Stability study protocol

5.2.2.1. Overview

The stability protocol developed for use in these studies was designed to include the following

information:

1) selection of batches,

2) number of batches,

3) container closure system,

4) sampling frequency,

5) sampling plan,

6) test storage conditions,

7) test specifications,

8) product specifications,

9) test procedure, and

10) statistical evaluation of the data.

A stability test summary sheet for the CP cream formulation used in these studies is shown in

Appendix 7.

5.2.2.2. Selection of batches

Specification requirements with respect to the selection of batches for stability testing have been

reported [217, 221]. According to the WHO [217] and the ICH [221] it is largely irrelevant

whether batches to be tested are of a pilot scale or large scale. However, the batches to be

sampled and tested should be representative of the manufacturing procedure used to produce the

product. Therefore the batches of CP cream that were subjected to stability testing were selected

from amongst the best prototype batches viz., Batches CP004, CP005 and CP006 (Table 4.2),

which were manufactured using the same procedure as described in Section 4.3.5. The size of

each prototype batch that was produced was 500 g.

173

5.2.2.3. Number of batches

Current international guidelines on stability evaluation of drug substances and/or drug products

recommend the testing of a minimum of three (3) batches [217, 221, 227]. The use of three (3)

batches may be a compromise between statistical and practical considerations [218, 227]. The

testing of fewer batches may not allow for a reliable statistical estimate of batch-to-batch

variability and the impact on product stability [218, 227]. However, practical considerations may

prevent the collection of large amounts of data, since the generation of excessive data may cause

stress on analytical and other facilities within a company [218, 227]. Testing of fewer batches

such as, for example, one (1) batch may be allowed for stable and well-established products

produced on a pilot scale [218]. As a consequence, one (1) batch of the three best CP cream

products viz., Batch CP004 was used in these studies.

5.2.2.4. Container-closure system

According to the current guidelines [217, 221, 227] stability testing should be conducted using

containers and closures that will be used for the marketing of the medicinal product. However, as

part of the conditions for the accelerated stability testing of a dosage form, the ICH guidelines

[221] consider the use of other packaging materials to be appropriate to generate supporting

stability data for a product produced at a pilot scale level. As a consequence, the CP cream

formulation was packed into 100 g glass ointment jars with tight-fitting closures. These

containers were selected for use since it was easy to measure formulation parameters such as

viscosity and pH directly from the ointment jar following storage of the samples for the requisite

times (Section 5.2.2.8).

5.2.2.5. Sampling frequency

Accelerated stability studies may be undertaken for a minimum of six (6) months [217, 221,

227]. During the six (6) month test period, the ICH suggests that samples should be removed at a

minimum of three (3) time points including the initial and final time points, for example, at 0, 3,

and 6 months [221]. The FDA [227] however sets the sampling frequency at a minimum of four

174

(4) time points, for example at 0, 2, 4 and 6, whereas the WHO [217] has set the sampling

frequency at a minimum of five (5) time points such as, for example, at 0, 1, 2 and 3 and 6

months (WHO). Since the objective of the current work was to determine whether the CP cream

formulation is a potentially stable system, stability studies were conducted for a period of one (1)

month and the sampling frequency in these studies was set at 1, 2, 3 and 4 weeks.

5.2.2.6. Sampling plan

Following completion of a sampling time plan, the total number of containers required for

stability testing must be determined [218]. The primary requirement for determination of sample

numbers is that the containers that are selected must represent the batch as a whole [218]. The

formulation to be tested was packed into five (5) different 100 g glass ointment jars, resulting in

the production of five (5) different test samples. The use of five (5) test samples was necessary to

facilitate sampling at the different time points of 0, 1, 2, 3, and 4 weeks such that at the

appropriate time, a jar could be removed without interfering with the remaining samples retained

in the stability oven.

5.2.2.7. Test storage conditions

Haynes [228] gathered climatic data in various cities worldwide and has proposed a formula for

the determination of the mean kinetic temperature (MKT) in these locations by taking into

consideration variations in storage temperature. As a consequence different countries around the

world have been distributed into four (4) different climatic zones, based on the prevailing annual

climatic conditions in those regions [217-219]. The Republic of South Africa (RSA) falls into

either Zones I or II [218, 225].

For countries located in Zones I and II accelerated stability studies for pharmaceutical products

should be conducted at 40 ± 2˚C and 75 ± 5% RH [217, 221, 227]. Due to lack of a controlled

humidity stability chamber in our laboratory, stability studies on the CP cream formulation were

conducted at 40 ± 2˚C and 25 ± 5% RH, which were the test storage conditions that could be

achieved with the facilities available in our laboratory. Whilst saturated salt chambers [229] may

175

be used to achieve the desired degree of humidity, the excessive salt required may affect the

stability of the dosage form and containers and closures or cause corrosion and therefore these

were not considered suitable for use.

5.2.2.8. Test specifications

The specifications for a stability test are designed to facilitate the selection of test parameters,

analytical procedures and proposed acceptance criteria for a specific product [221]. Stability

studies require that parameters of a product that may be susceptible to change during storage and

that are more than likely to influence the quality, safety and/or efficacy of a product be tested

[221].

For topical formulations such as creams, attributes that should be investigated in stability studies

may include the organoleptic properties and homogeneity of the dosage form, pH, viscosity,

particle size distribution, strength, weight loss and API release rate [227, 230]. For the purpose

of these studies, the CP cream formulation was evaluated in terms of its organoleptic attributes,

CP content, consistency, pH and in vitro release rate of CP. The analytical procedures and

acceptance criteria used in these studies are discussed in Sections 5.2.2.9 and 5.2.2.10,

respectively, vide infra.

5.2.2.9. Product specifications

The test product specifications for the qualitative and quantitative parameters tested in these

studies are listed in Table 5.1 and Table 5.2 respectively. The objective of these stability studies

was to determine whether or not the organoleptic properties (Table 5.1) and the physico-

chemical parameters (Table 5.2) of the test formulation (Batch CP004) would remain within the

specifications set prior to testing following exposure of the product to the stability test conditions

used in these studies (Section 5.2.2.7). These specifications were determined for Batch CP004

based on the data generated at the time of manufacture of that batch (Section 4.3.8).

176

Table 5.1. Specifications for qualitative parameters of the test cream formulation

Product parameter Specification Appearance Shiny Colour White Odour Odourless Signs of physical instability None

Table 5.2. Specifications for quantitative parameters of the test cream formulation

Product parameter Specification CP content 90-115 % Intrinsic viscosity 13-18 KcP Intrinsic pH 4-6 In vitro release rate (flux) 5-8 µg/cm2/hr1/2

5.2.2.10. Methodology

5.2.2.10.1. Test procedure

Stability studies were conducted using a Model L.T.I.E. Labcon low temperature incubator

(Labcon (Pty) Ltd, Krugerdorp, Gauteng, RSA). A week prior to use, the incubator was allowed

to equilibrate to the temperature and relative humidity conditions set for these studies and these

test conditions were monitored using a Thermo-Hygro pen (Beijing Gaupu Automation Control

Co., Ltd, Beijing, China). Four (4) containers containing 100 g of cream were placed in the

equilibrated incubator and at weeks 1, 2, 3 and 4 a 100 g sample was removed from the incubator

and immediately analysed in terms of the following test parameters:

a) organoleptic appeal,

b) CP content,

c) viscosity,

d) pH, and

e) CP release rate.

The data generated in these tests were subjected to statistical analysis where appropriate.

177

5.2.2.10.2. Organoleptic appeal

The appearance, odour and colour of the cream in each container was visually evaluated.

Samples were also assessed for any signs of physical instability, such as for example, cracking

and/or bleeding of the cream base from the container.

5.2.2.10.3. CP content

The CP content of the cream was evaluated using the sample preparation procedure described in

Section 2.4.6.2 at each sampling time and a maximum of three (3) cream samples were used

obtain an average value of CP. CP content was determined using a validated HPLC method with

UV detection at 240 nm (Chapter 2). These samples were assessed immediately after removing

the cream sample container from the incubator.

5.2.2.10.4. Intrinsic viscosity

The apparent intrinsic viscosity of the creams was measured as described in Section 3.3.2.3.2 at

each sampling time and a maximum of three (3) readings were taken to obtain an average

apparent intrinsic viscosity value. Intrinsic viscosity readings were taken once, immediately after

removing the cream sample from the incubator.

5.2.2.10.5. pH-determination

The apparent pH was determined as described in Section 4.3.1.4 at each sampling time and a

maximum of three (3) readings were taken to obtain an average apparent pH value. The apparent

pH readings were taken once, immediately after removing the cream sample from the incubator.

5.2.2.10.6. In vitro release test

The in vitro release rate of CP was assessed using the in vitro release test conditions described in

Table 3.5, and were assessed following removal of the sample at each sampling time. A

178

maximum of six (6) Franz glass diffusion cells were used to determine the average in vitro

release rate of CP (flux). In vitro release rate studies were conducted immediately after removing

the cream sample from the incubator.

5.2.2.11. Statistical evaluation

In addition to describing how a stability study is to be designed and undertaken, the stability

study protocol must also establish the statistical procedures to be used to analyse the stability

data generated in the stability studies [227]. Stability data evaluation methods proposed in the

current international guidelines [217, 221, 227] and in some published reports [218, 231-233] are

intended to establish an expiration dating period during which the average strength of a batch of

drug product may be expected to remain within the specifications for that product. The

guidelines [217, 221, 227] and reports [231, 232] do not specify the statistical methods to be

used to predict the stability of a drug product in terms of the physical attributes of a formulation

such as viscosity, pH and in vitro release rates as were determined in these studies.

The purpose of conducting these studies was to assess whether or not exposing the CP cream

formulation to the test conditions reported in Section 5.2.2.7 would cause any relevant and/or

significant changes to the organoleptic and physico-chemical properties of the cream. In terms of

the qualitative tests, i.e. organoleptic properties, the formulation was considered to be stable if no

changes were observed in terms of appearance, colour and odour of the product and if signs of

physical instability, such as cracking and/or bleeding were absent during the test period. The

organoleptic appeal was assessed qualitatively.

The physico-chemical attributes of the formulation were evaluated quantitatively. The

formulation was considered to be stable if there was no statistical relevant and/or significant

change between the test results of the 100 g unit assessed immediately after manufacture (time 0

weeks) and the 100 g units stored at accelerated test conditions and analysed at 1, 2, 3 and 4

weeks after storage. In order to assess whether or not these changes were statistically relevant

and/or significant, the statistical procedure described by Timm et al., [97] as described in Section

2.4.7.2 was used.

179

The responses used to calculate the confidence intervals as described by Timm et al., [97] were

the CP content, apparent intrinsic viscosity, apparent pH and CP in vitro release rates. The

percentage difference in response between the initial sample (at time 0 weeks) and stored

samples (at times 1, 2, 3 and 4 weeks) were calculated and used to construct a 90% C.I.. The data

are reported as confidence intervals depicting the percentage change from the initial reading for

CP content in the formulation (n =3), intrinsic viscosity (n = 3) and apparent pH (n = 3) of the

product and CP in vitro release rates from the formulation (n = 6).

In addition, similarities or differences in the in vitro release rates of CP from each of the stability

samples vs. the in vitro release rate of CP from the reference sample were evaluated using a non-

parametric test, which was described and reported in Section 3.2.11.

5.2.3. Results and discussion

5.2.3.1. Qualitative analysis

There were no noticeable changes in the organoleptic properties of Batch CP004 in terms of

appearance, colour and odour (Table 4.1) over the entire stability test period. In addition the

formulation did not show any signs of physical instability such as phase separation or cracking

and bleeding of the cream base from the container at any of the sampling times. Based on the

organoleptic data observed in these studies the CP cream formulation may be considered to be

potentially stable.

5.2.3.2. Quantitative analysis

The stability data for Batch CP004 evaluated in terms of CP content, intrinsic viscosity, intrinsic

pH and in vitro release rate (flux) of CP from the cream formulation, generated over the entire

stability test period are summarized in Table 5.3.

180

Table 5.3. Stability data generated for Batch CP004 after a four (4) week test period

Stability test parameters Sampling time

(week) CP content (%) Intrinsic

viscosity (KcP) Intrinsic pH Flux

(µg/cm2/h1/2) 0 111.7 ± 1.906 15.13 ± 0.7023 5.11 ± 0.00577 7.532 ± 0.8056 1 110.4 ± 2.871 16.27 ± 1.115 5.15 ± 0.0265 7.800 ± 0.6016 2 109.6 ± 2.474 17.17 ± 0.4163 5.21 ± 0.0568 7.480 ± 0.3761 3 103.7 ± 4.418 17.70 ± 0.6080 5.17 ± 0.0550 7.626 ± 0.08207 4 101.7 ± 4.073 17.23 ± 0.1528 5.25 ± 0.0608 6.727 ± 1.468

5.2.3.2.1. CP content

The data shown in Table 5.3 indicate a progressive decrease in the amount of CP in the

formation as the time of exposure of the formulation to the stability test conditions increased.

This may be due to degradation of the CP in the formulation and the longer the formulation

remained exposed to the test conditions, the greater is the amount of CP that degrades. However,

studies to establish whether or not the degradation of CP in the cream base occurred were not

undertaken. Nevertheless, despite the progressive decrease in the amount of CP in Batch CP004

after exposure to stability test conditions, the CP content remained within the specifications

established for the formulation (Table 5.2) at all sampling times.

The CP content data generated in these tests were subjected to a statistical analysis (Section

2.4.7.2) and the results obtained from this test are shown in Figure 5.1. These results indicate that

the percentage change from the initial amount of CP in the formulation after storage at 40 ± 2˚C

and 25 ± 5% RH for 1, 2, 3, and 4 weeks is not statistically significant or relevant. Consequently,

the CP cream can be considered as stable for the time period under investigation.

181

-20 -10 0 10

% Change from initial CP content

20

1 week at 40 C25% RH

o

2 weeks at 40 C25% RH

o


o


o

Figure 5.1. Effects of stability test conditions on CP content of the cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks.


The data shown in Table 5.3 showing the apparent intrinsic viscosity indicate that the apparent

intrinsic viscosity of Batch CP004 remained within the established specifications (Table 4.2)

over the stability test period.

There seem to be a noticeable increase in the apparent intrinsic viscosity value of all the test

samples when compared to the viscosity of the cream sample observed at time 0 weeks. The

apparent intrinsic viscosity data generated in these tests were subjected to a statistical analysis

(Section 2.4.7.2) to determine whether or not the increase was significant and/or relevant and the

results obtained from this test are illustrated in Figure 5.2.

These data reveal that the percentage change from the initial apparent intrinsic viscosity of the

CP cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3, and 4 weeks is not

182

statistically significant or relevant. Based on these results, the CP cream formulation is

considered stable for the 1 month test period under investigation.

-20 -10 0 10

% Change from initial intrinsic viscosity

20


o


o


o


o

Figure 5.2. Effects of stability test conditions on the apparent intrinsic viscosity of the CP cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks.

5.2.3.4. Apparent intrinsic pH

The data shown in Table 5.3 showing the apparent pH of the cream samples indicate that the

apparent pH of the cream formulation remained within the established specifications (Table 4.2)

over the entire test period. It appears that the apparent pH of the formulation increased slightly

after exposure to the stability test conditions and this increase was proportional to the time of

exposure.

In order to determine whether or not the increase in the apparent intrinsic pH of the formulation

was statistically significant and/or relevant, the apparent intrinsic pH data were subjected to a

statistical analysis (Section 2.4.7.2) and the results generated from this test are illustrated in

Figure 5.3.

183

These data reveal that the percentage change from the initial apparent pH of the CP cream

formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3, and 4 weeks is not statistically

significant or relevant. As a consequence, these results reveal that the pH of the product is still

within the specified pH range for topical formulations [193, 194] and within the range set for this

product (Table 5.2). The CP cream formulation can be considered stable in terms of maintaining

the pH within the specified limits during the time period the formulation was tested.

-20 -10 0 10

% Change from initial pH

20


o


o


o


o

Figure 5.3. Effects of stability test conditions on the apparent intrinsic pH of the cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks. . 5.2.3.5. In vitro release rate testing

The data shown in Table 5.3 for the in vitro release rate (flux) of CP from the cream reveal that

the in vitro release rate (flux value) of CP from the CP cream remains relatively constant but

shows a slight decrease in flux for the 4 weeks sample that appears significant. Nevertheless, all

in vitro release rates (flux values) calculated in these studies fall within the in vitro release rate

specifications established for the cream formulation (Table 5.3). There appears to be a

184

correlation between this slight decrease in flux and the slight increase in the viscosity of the

formulation following four weeks of storage.

In order to determine whether or not changes between the in vitro release rates of CP release

from cream samples are statistically significant and/or relevant, the in vitro release rate of CP

data were subjected to a statistical analysis (Section 2.4.7.2) and the data generated are shown in

Figure 5.4.

These data indicate that the percentage change from the initial CP in vitro release rate of CP

release from the cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for weeks 1, 2, 3,

and 4 is not statistically significant or relevant. These results indicate that the cream formulation

is stable over the period tested.

-20 -10 0 10

% Change from initial release ratein vitro

20


o


o


o


o

Figure 5.4. Effects of stability test conditions on the in vitro release rates of CP from the cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3 and 4 weeks

185

The data collected during in vitro release rate studies were also used to generate in vitro release

rate profiles for CP and the data are plotted in Figure 5.5. These data are plotted as the

cumulative amount of CP released per unit area (Q) vs. square root of time (t1/2). From these

results, it is evident that there is no difference in the in vitro release rate profiles of CP from the

cream samples tested over the entire test period.

0

10

20

30

40

50

60

70

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )

week 0 week 1 week 2 week 3 week 4

Figure 5.5. In vitro release rate profiles for CP from CP cream samples at times 0, 1, 2, 3 and 4 weeks after storage

In order to establish whether the apparent similarity in the in vitro release rate profiles for CP

following storage of creams and the cream tested immediately after manufacture was real the

data were tested using non-parametric statistical analysis (Section 3.2.11).

The lower limit (L.L.) and upper limit (U.L.) of a confidence interval (C.I.) calculated as

described in Section 3.2.11 (Chapter 3) with the cream sample at week 0 (reference) and the

cream sample at weeks 1, 2, 3 or 4 (test) are summarised in Table 5.4. If the in vitro release rate

profile of CP release from the reference sample is to be considered equivalent to the in vitro

186

release rate profile for CP release from the test sample then the limits for the C.I. calculated from

experimental data should lie within the 75%-133% limits [49].

Table 5.4. The lower limit (L.L.) and upper limit (U.L.) of the confidence intervals (C.I.) calculated using cream sample at week 0 (reference) and cream samples at weeks 1, 2, 3 and 4 (test)

Confidence intervals (C.I.) L.L. (%) U.L. (%) C.I. limits 75.00 133.0 Week 0 vs. week 1 93.97 114.0 Week 0 vs. week 2 89.43 110.9 Week 0 vs. week 3 93.88 110.0 Week 0 vs. week 4 86.02 108.1

From the results summarised in Table 5.4 it is clear that there is no statistically significant

difference in the in vitro rate of release of CP release from creams subjected to stability test

conditions. These data are consistent with the results generated using the statistical test described

by Timm et al., [97] (Figure 5.5), which reveal that the in vitro release rates of CP release from

the tested cream samples at weeks 1, 2, 3 and 4 are equivalent to the in vitro release rate of CP

release from the cream sample tested at week 0. Consequently Batch CP004 can be considered

stable over the time period the formulation was tested

5.3. CONCLUSIONS

The stability of the CP cream formulation developed, assessed and reported in Chapter 4 has

been evaluated using accelerated stability test conditions. The initial step taken in these studies

was to develop a stability study protocol using currently published international guidelines for

the stability testing of drug substances and drug products. Essentially, the stability protocol used

in these studies was designed to facilitate the selection of batches, number of batches, container

closure system, sampling frequency, sampling plan, test storage conditions, test specifications,

test procedure and stability data evaluation procedure.

One of the best prototype formulations, Batch CP004, was selected for use in stability testing

studies. Based on the recommendations in the guideline documents a stability study on a single

batch of the generic formulation was considered to be adequate for the purposes of these

187

investigations. The formulation was packed into four (4) 100 g glass ointment jars, to facilitate

ease of sampling at the four (4) time points, 1, 2, 3, and 4 weeks.

The stability study was conducted under accelerated conditions of 40 ± 2˚C and 25 ± 5% RH.

Internal guidelines on stability studies specify that accelerated stability studies be conducted at

40 ± 2˚C and 75 ± 5% RH. However, these conditions were slightly modified due to the

availability of facilities in our laboratory. The main objective of these studies was to determine

whether or not the modified accelerated stability conditions would have an effect on specific

formulation attributes such as organoleptic appeal, CP content, apparent intrinsic viscosity,

apparent pH and CP in vitro release rates.

Sample were removed from the storage incubator at weeks 1, 2, 3 and 4 and immediately tested

for organoleptic appeal, CP content, apparent intrinsic viscosity, apparent intrinsic pH and the

CP in vitro release rates.

Organoleptic data obtained from these studies were evaluated qualitatively, whereas quantitative

data were generated for CP content, apparent intrinsic viscosity and pH and CP in vitro release

rates. The data generated at each sample time were assessed using a statistical test procedure to

determine whether a relevant and/or significant change occurred in the product following storage

for up to four weeks as compared to that evaluated immediately after manufacture.

The original appearance of the cream formulation did not change and the formulation did not

show signs of physical instability, such as phase separation or cracking and bleeding of the

cream base from the container at any of the sampling times. Quantitative analysis revealed that

the percentage change from the initial product specifications such as CP content, apparent

intrinsic viscosity and pH and CP in vitro release rates was in all cases not statistically significant

or relevant.

Based on the data generated in this pilot stability study, the CP cream formulation appears to be

stable. However, it will be necessary to conduct additional studies of the cream formulation in

conventional cream packaging and a stability chamber using the established accelerated stability

188

test conditions for longer periods to fully characterize the short-term and long-term stability of

this formulation. Furthermore, it will be essential in future studies to conduct stability studies on

the innovator product (Dermovate® cream) and compare the data generated from these studies to

those obtained for the generic formulation.

It is evident that the formulation development process was successful in producing a potentially

stable CP cream product and these studies could form the basis for further studies on the

formulation that has been developed.

189

CHAPTER SIX

CONCLUSIONS

Clobetasol 17-propionate (CP) has been labelled as a class I topical corticosteroid and is

currently one of the most potent topical corticosteroids available on the market. CP is used for

the short-term treatment of inflammatory and pruritic manifestations of moderate-to-severe

glucocorticoid-responsive dermatoses. CP has physicochemical and pharmacological properties

such as a relative high lipophilicity and local anti-inflammatory activity which make it an ideal

candidate for incorporation into topical semi-solid formulations. The objective of this research

project was to develop and characterize a CP topical cream formulation containing 0.05% w/w of

the drug.

Prior to initiating formulation development studies, it is vital that a suitable analytical method is

developed and validated for the quantitation of drug and the characterization of dosage forms

during the development and assessment process. A major difficulty often encountered in the

analysis of semi-solid dosage forms is interference due to formulation adjuvants and

preservatives that are usually present in what are relatively complex formulations. RP-HPLC is a

commonly used, powerful and reliable analytical tool that can be used for the in vitro analysis of

formulations of a complex nature, such as for example creams, gels and ointments.

HPLC not only provides separation and quantitative data but also has the ability to eliminate

almost all interference challenges. The initial phases of this project entailed the development and

validation of a suitable RP-HPLC method that was suitable for the quantitative analysis of CP in

cream formulations and CP release during in vitro release studies. Separation of CP and the

internal standard (BV) was achieved on a Nova-Pak® C18 cartridge column (3.9 x 150 mm, 4µm)

using a mobile phase consisting of acetonitrile-water (50:50) at a flow rate of 1.0ml/min with UV

detection at 240 nm. The retention time of BV and CP were 5.6 and 8.2 min respectively and the

total run time for analysis was 10 min.

190

The RP-HPLC analytical method was validated according to USP [23], FDA [98] and ICH [86]

guidelines. The analytical method was found to be linear, precise, accurate, selective and

sensitive and suitable for the intended purpose of CP quantitation. CP was found to be stable in

the mobile phase following storage at 4˚C for a maximum of 14 days, and thus all calibration

standards, which were made in the mobile phase, were stored at 4˚C and used within a period of

14 days, after which fresh calibration standards were prepared.

In vitro dissolution testing of solid oral dosage forms is a well-established technique used to

guide the formulation development process, assess product quality and ensure batch-to-batch

uniformity. However, as far as semi-solid formulations are concerned, only quality control tests,

such as for example the determination of solubility, particle size and size distribution and

crystalline form of an API, and evaluation of the intrinsic viscosity and homogeneity of a final

product has been traditionally used to provide reasonable evidence of consistent product

manufacture and performance. The main drawback with these quality control tests is that they

provide little information about drug release characteristics of a product or the effects of

processing and manufacturing variables on the performance of a finished product.

In recent years, some international guidelines [49] have recommended the use of in vitro release

test methods to determine if the diffusional rate of release of a drug from a formulation is the

same following any post-approval formulation changes, as it was prior to the changes. Such tests

can be used to detect the effects of changes in formulation composition on the rate of release of

an API from a dosage form in which that API is suspended and/or dissolved and can therefore be

used to ensure that the manufacture of semi-solid products is consistent. Therefore such tests

may fulfil a similar role as dissolution testing does for tablets and capsules in evaluating topical

semi-solid dosage forms.

In contrast to dissolution testing, for which official test methods have been developed and

reported for use in in vitro dissolution studies of solid oral dosage forms, there are currently no

official guidelines or requirements for the performance evaluation of drug release from semi-

solid dosage forms. The onus is left to a formulation scientist to develop and validate specific in

vitro release methods for the assessment of drug release from semi-solid products. Therefore, it

191

was essential to develop and validate an in vitro release test method for use in conjunction with

traditional quality control tests to determine the quality and consistency of CP topical cream

formulations that were to be manufactured during product development studies. The in vitro

release method was also used to characterize formulations and evaluate batch-to-batch

consistency, and in assessing the impact of storage at elevated temperatures on product

performance.

The in vitro release test method development studies entailed the selection of a suitable diffusion

cell system, appropriate sampling times, receptor medium, temperature of the receptor medium

and appropriate synthetic membranes. In addition the effects of sample application and sample

occlusion or non-occlusion on in vitro release of CP from semi-solid dosage forms were also

investigated. A modified Franz glass diffusion cell system was selected and used in these studies

since it has the most potential for use as a standardized test system that may be adapted for use as

a compendial method.

The in vitro release of CP from 0.05% w/w cream formulations was tested using a receptor

medium consisting of a binary mixture of water:propylene glycol (50:50) and a 0.025 µm

nitrocellulose membrane. The temperature of the test system was set at 32° ± 0.5˚C to

approximate the usual surface temperature of the skin and a 2 x 2 mm star head magnetic stirrer

was used to agitate the receptor medium.

About 300 mg of the semi-solid preparation that corresponded to an infinite dose was applied

uniformly to the membrane and the formulation was occluded for the duration of testing, to

prevent solvent evaporation and compositional changes to the formulation. Samples of the

receptor medium were withdrawn at 2, 4, 8, 12, 24, 48 and 72 hours in order to generate a

satisfactory CP release profile and to characterize the release of CP from semi-solid topical

formulations.

The in vitro release test method was validated using published protocols. Method validation

studies involved assessing the ability of the test method to detect the effects of changes in

192

formulation characteristics on CP release rates. The effects of changes of strength, composition

and intrinsic viscosity of topical formulations on in vitro release of CP were investigated.

It was determined that the in vitro release method was able to detect the effects on CP release

rates of changes in formulation in which CP is suspended and/or dissolved. The in vitro release

test was used in conjunction with traditional quality control tests to determine the quality and

consistency of CP topical formulations and the associated release characteristics of CP from the

product during formulation development and assessment studies.

In generic formulation development studies it is important to characterize the innovator product

to generate data, which can be used as a benchmark for the development of the generic

formulation. The initial experiments in these studies were designed to facilitate the

characterization of the innovator formulation, Dermovate® cream. CP content, apparent intrinsic

viscosity, apparent pH and in vitro release rates of CP from the formulation were determined and

the data obtained from these studies were used as reference specifications for the development of

a generic formulation.

Preliminary studies were designed to facilitate the manufacture of a prototype generic

formulation that showed no signs of physical instability, such as, for example, cracking,

creaming, phase inversion and/or bleeding of the cream base from the container, immediately

after the manufacture as well as twenty-four (24) hours after manufacture and storage at room

temperature (22˚C). As a consequence, a prototype formulation, Batch CP001 was

extemporaneously manufactured with Estol® 1474 as the primary emulsifier. Although Batch

CP001 showed no signs of physical instability immediately after manufacture, the formulation

had cracked twenty-four (24) hours after manufacture and storage at room temperature (22˚C)

and was therefore not considered for further development.

Three additional prototype formulations containing commercially available mixed emulgents,

Ritapro® 200 (Batch CP002), Emulcire® 61 WL (Batch CP003) and Gelot® 64 (Batch CP004)

were manufactured. All three formulations were found to be physically stable immediately after

193

manufacture as well as twenty-four (24) hours after manufacture and storage at room

temperature (22˚C) and were therefore considered for further development.

Following the manufacture of stable batches, the next phase of the formulation development

studies was to investigate the in vitro release rate of CP release from Batches CP002, CP003 and

CP004 and Dermovate® cream. These studies were conducted in order to identify a prototype

batch with a similar in vitro release rate profile for CP similar to that of Dermovate® cream.

The similarities or differences between the in vitro release rates of CP release from the prototype

cream formulations and Dermovate® cream were statistically determined using a non-parametric

test recommended by the FDA [49]. It was found that only the in vitro release rate of CP release

from Batch CP004 was found to be equivalent to the in vitro release rate of CP release from the

innovator formulation. Consequently, only Batch CP004 was developed and investigated further.

Two additional batches, Batches CP005 and CP006, having similar a composition to Batch

CP004, were manufactured using the same manufacturing procedure as that used to produce

Batch CP004. The three formulations were then characterised for CP content, apparent intrinsic

viscosity, pH and in vitro release rate and the data generated from these studies were compared

to the data generated for the innovator product.

The CP content of all generic formulations and the Dermovate® cream were comparable and fell

within the USP limits for CP content in topical semi-solid formulations. In addition, the apparent

pH values of Batches CP004, CP005 and CP006 were comparable to that of the innovator

product and fell within the pH range of healthy human skin.

The apparent intrinsic viscosity of Dermovate® cream was found to be almost three (3) times

greater than that of generic cream batches and the difference in intrinsic viscosity of Dermovate®

cream was attributed to the physicochemical characteristics of the mixed emulsifiers used in the

innovator product and to manufacturing process variables, such as for example speed of

homogenization speed and mode of cooling.

194

Despite the differences in apparent intrinsic viscosity the in vitro release rate of CP from Batches

CP004, CP005 and CP006 was found to be statistically similar to the in vitro release rate of CP

from the innovator product. It was concluded that the behaviour of the mixed emulsifier in both

the commercially available cream and the extemporaneously manufactured creams, and not only

the apparent intrinsic viscosity of the formulations, had e an impact on the in vitro release rate of

CP from these formulations.

Based on the comparative data generated for the generic cream and those for the innovator

formulation, the formulation development process was considered to have been successful.

Despite the fact that the apparent intrinsic viscosity of the generic formulations was significantly

different to that of the Dermovate® cream, Batches CP004, CP005 and CP006 were found to be

equivalent to the innovator cream product in terms of other in vitro performance characteristics

such as CP content, pH and in vitro release rates.

Following the development of batches of cream that were considered equivalent to the innovator

formulation in terms of in vitro performance characteristics, three 500 g batches were produced

for use in accelerated stability studies. The stability studies aimed to determine whether or not

the formulation development process was successful in producing a stable CP cream product.

Therefore the stability study was designed to facilitate the selection of batches, number of

batches, container closure system, sampling frequency, sampling plan, test storage conditions,

test specifications, product specifications, test procedure and stability data evaluation procedure.

Batch CP004 was selected for testing and the formulation was packed into five (5) 100 g glass

ointment jars, to facilitate ease of sampling at the five (5) time points, 0, 1, 2, 3, and 4 weeks,

and a stability test was conducted under conditions of 40 ± 2˚C and 25 ± 5% RH in an incubator.

Cream samples were tested for qualitative and quantitative attributes such as organoleptic appeal,

CP content, apparent intrinsic viscosity, apparent pH and in vitro release rate. The product

specifications for all parameters were set prior to initiating stability testing and qualitative data

were generated by visual observation, whereas quantitative data were analysed and evaluated

195

using statistical testing. The cream samples were found to be physically stable over the entire test

period and the samples did not change in appearance, colour or odour.

Although, there appeared to be a progressive decrease in CP content of the cream as the time of

cream exposure to the stability test conditions increased, the decrease was not statistically

significant or relevant. The apparent intrinsic viscosity and pH values of the formulation

increased slightly over the test period, but again these increases were found to be neither

statistically significant nor relevant.

The in vitro release rate of CP from the cream formulation appeared to remain constant, after 4

weeks a decrease from the initial in vitro release rate of CP was observed. Nevertheless,

statistical analyses indicated that the changes in in vitro release rates were not significant or

relevant.

The data generated during accelerated stability studies indicate that the CP cream formulation

manufactured is potentially stable. However, it will be necessary to conduct additional studies on

the cream formulation stored in conventional cream packaging and using a stability chamber

with the ability to create established accelerated stability test conditions. Furthermore the product

would have to be tested for longer periods to fully characterize the short-term and long-term

stability of this formulation. In addition, stability studies would need to be conducted on the

innovator product (Dermovate® cream) in order to provide comparative data for the assessment

of the generic formulation.

Future studies would entail an investigation of the effects of formulation and process variables

on in vitro performance of the dosage forms. The apparent intrinsic viscosity of the formulations

must be increased to match the viscosity of the innovator product. Intrinsic viscosity invariably

affects drug release from a semi-solid formulation and although the viscosity does not appear to

affect the in vitro release of CP from the semi-solid formulations developed in these studies, the

viscosity may have a significant effect when testing the products in vivo.

196

The formulation appears stable at high temperatures following pilot-scale manufacture and

therefore scale-up of the process is necessary and challenges associated with formulation scale-

up require investigation.

In vivo studies could be conducted using the the human skin blanching assay [7] to determine

whether the generic formulation developed in these studies is bioequivalent to the innovator

product. In vitro release rate testing with rate limiting membranes, such as for example a silastic

membrane or membranes derived from animal models, may assist in the development of a test

model to establish an in vitro in vivo correlation (IVIVC) for CP release from topical

formulations.

A stable cream formulation has been developed and assessed and the data generated reveal that

the dosage form behaves similarly to Dermovate® during in vitro studies. The results obtained

provide a suitable platform from which a bioequivalent CP semi-solid cream can be developed

and commercialised.

197

APPENDIX ONE

EXAMPLE OF THE CALCULATION OF A 95% CONFIDENCE INTERVAL FOR THE

RATIO OF T/R RELEASE RATES

A. Release rates (slopes) for reference system (R) and test system (T) Cell No. Test Reference

1 5.4050 5.1894 2 6.0153 4.3467 3 6.1577 5.4699 4 6.2852 5.9206 5 5.3946 5.2937 6 5.3641 4.3055

B. Computation of 36 individual T/R ratios (R) slopes (T) 5.1894 4.3467 5.4699 5.9206 5.2937 4.30555.4050 1.0415 1.2435 0.9881 0.9129 1.0210 1.2554 6.0153 1.1592 1.3839 1.0997 1.0160 1.1363 1.3971 6.1577 1.1866 1.4166 1.1257 1.0400 1.1632 1.4302 6.2852 1.2112 1.4460 1.1491 1.0616 1.1873 1.4598 5.3946 1.0395 1.2411 0.9862 0.9112 1.0191 1.2530 5.3641 1.0337 1.2341 0.9807 0.9060 1.0133 1.2459 C. Arrangement T/R ratios from the lowest to highest (counting from left to right) 0.9060 0.9112 0.9129 0.9807 0.9862 0.9881 1.0133 1.01601.0191 1.0210 1.0337 1.0395 1.0400 1.0415 1.0616 1.09971.1257 1.1363 1.1491 1.1592 1.1632 1.1866 1.1873 1.21121.2341 1.2411 1.2435 1.2459 1.2530 1.2554 1.3839 1.39711.4166 1.4302 1.4460 1.4598

D. The eighth T/R ratio converted to % is 101.6% and the twenty-ninth T/R ratio converted to % is 125.3%, which lies between 75% and 133.3%. The test product is therefore equivalent to the reference product

198

APPENDIX TWO

QUALITATIVE COMPOSITION OF DERMOVATE® CREAM

Qualitative composition of Dermovate® cream [196]

Excipients Clobetasol propionate Chrorocresol Cetostearyl alcohol Glyceryl monostearate Arlacel® 165 Beeswax substitute Propylene glycol Sodium citrate Citric acid Purified water

199

APPENDIX THREE

EXCIPIENTS USED IN CP CREAM FORMULATION DEVELOPMENT STUDIES

Excipients Trade name Manufacturer/Donor Clobetasol 17-propionate Clobetasol propionate Symbiotec Pharmalab P.V.T. Ltd (Pigdamber, Maharastra, India Propylene glycol propane-1,2-diol Merck Chemicals (pty) Ltd, Darmstadt, Germany Sodium citrate N/A Aspen Pharmacare, Port Elizabeth, Eastern Cape, SA Citric acid N/A Aspen Pharmacare, Port Elizabeth, Eastern Cape, SA Glyceryl monostearate Geleol® Gattefosse SAS, Saint-Priest Cedex, France Cetostearyl alcohol N/A Aspen Pharmacare, Port Elizabeth, Eastern Cape, SA White beeswax BP N/A Croda Chemicals (SA) (Pty) Ltd, Johannesburg, Gauteng, SA Chlorocresol N/A Aspen Pharmacare, Port Elizabeth, Eastern Cape, SA Glyceryl stearate Estol® 1474 Uniqema (Pty) Ltd, Bryanstone, Gauteng, SA Stearyl alcohol and steareth-20 Ritapro® 200 Rita, Crystal Lake, IL, USA Cetyl alcohol, ceteth-20 and steareth-20 Emulcire® 61 WL Gattefosse SAS, Saint-Priest Cedex, France Glyceryl stearate and PEG-75 stearate Gelot® 64 Gattefosse SAS, Saint-Priest Cedex, France

200

APPENDIX FOUR

QUALITATIVE AND QUANTITATIVE FORMULAE USED TO MANUFACTURE CP CREAM

FORMULATIONS IN THESE STUDIES

Material Composition (% w/w) Clobetasol propionate 0.05000 Propylene glycol 44.50 Sodium citrate 0.05000 Citric acid 0.05000 Glyceryl monostearate A/S 5.000 Cetostearyl alcohol 4.000 White wax (beeswax bleeched) 0.6000 Chlorocresol 0.0750 Glyceryl monostearate SE 1.000 Propylene glycol 7.000 Propylene glycol 2.675 Purified water 35.00

201

APPENDIX FIVE

BATCH SUMMARY REPORTS

202

RHODES UNIVERSITY, Faculty of Pharmacy, Grahamstown, SOUTH AFRICA

BATCH SUMMARY REPORT

Formulator: Kasongo Wa Kasongo Product: Clobetasol 17-propionate cream Batch ID: CP001 Batch Size: 200 g Manufacturing Date: 04-04-2006 Manual mixing time: 10 min High speed homogenization time: 5 min Formula

Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.1023 RM000150 Propylene glycol 44.50 89.04 RM000181 Sodium citrate 0.05000 0.1019 RM000183 Citric acid 0.05000 0.1026 RM000185 Geleol® pastilles 5.000 10.05 RM000182 Cetostearyl alcohol 4.000 8.053 RM000184 White beeswax BP 0.6000 1.020 RM000142 Chlorocresol 0.07500 0.1508 RM000186 Estol® 1474 1.000 2.005 RM000160 Propylene glycol 7.000 14.10 RM000181 Propylene glycol 2.675 5.350 RM000181 Purified water 35.00 70.07 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated:

In vitro release rate Comments / Observations Not evaluated

Mean SD CP content Not evaluated Not evaluatedViscosity Not evaluated Not evaluatedpH Not evaluated Not evaluated

• Shiny, white, odourless cream was produced

• The cream was physically stable immediately after manufacture

• The cream cracked 24 hours after manufacture

• Physico-chemical attributes of the cream were not evaluated

203




Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.1033 RM000150 Propylene glycol 44.50 89.10 RM000181 Sodium citrate 0.05000 0.1055 RM000183 Citric acid 0.05000 0.1008 RM000185 Geleol® pastilles 5.000 10.07 RM000182 Cetostearyl alcohol 4.000 8.071 RM000184 White beeswax BP 0.6000 1.218 RM000142 Chlorocresol 0.07500 0.1527 RM000186 Ritapro® 200 1.000 2.019 RM000140 Propylene glycol 7.000 14.01 RM000181 Propylene glycol 2.675 5.450 RM000181 Purified water 35.00 70.03 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated:

In vitro release rate Comments / Observations

01020304050607080

0 2 4 6 8 10

Time (h)0.5

Q (µ

g/cm

2 )


Mean SD CP content (% ) 104.6 3.722 Viscosity (KcP) 23.07 0.9741 pH 5.20 0.0152



• The cream remained physically stable 24 hours after manufacture

• Physico-chemical attributes of the cream were evaluated

204




Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.1018 RM000150 Propylene glycol 44.50 89.00 RM000181 Sodium citrate 0.05000 0.1019 RM000183 Citric acid 0.05000 0.1026 RM000185 Geleol® pastilles 5.000 10.05 RM000182 Cetostearyl alcohol 4.000 8.053 RM000184 White beeswax BP 0.6000 1.220 RM000142 Chlorocresol 0.07500 0.1528 RM000186 Emulcire® 61 WL 1.000 2.005 RM000178 Propylene glycol 7.000 14.00 RM000181 Propylene glycol 2.675 5.350 RM000181 Purified water 35.00 70.00 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated:


01020

30405060

7080

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )


Mean SD CP content 113.1 2.605 Viscosity 11.43 0.05774 pH 5.15 0.0264





205




Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.2538 RM000150 Propylene glycol 44.50 222.8 RM000181 Sodium citrate 0.05000 0.2596 RM000183 Citric acid 0.05000 0.2548 RM000185 Geleol® pastilles 5.000 25.01 RM000182 Cetostearyl alcohol 4.000 20.13 RM000184 White beeswax BP 0.6000 3.01 RM000142 Chlorocresol 0.07500 0.3759 RM000142 Gelot® 64 1.000 5.100 RM000177 Propylene glycol 7.000 35.44 RM000181 Propylene glycol 2.675 13.68 RM000181 Purified water 35.00 175.0 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated:


0

10

20

30

40

50

60

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )







206




Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.2524 RM000150 Propylene glycol 44.50 222.8 RM000181 Sodium citrate 0.05000 0.2539 RM000183 Citric acid 0.05000 0.2533 RM000185 Geleol® pastilles 5.000 25.02 RM000182 Cetostearyl alcohol 4.000 20.07 RM000184 White beeswax BP 0.6000 3.02 RM000142 Chlorocresol 0.07500 0.3778 RM000142 Gelot® 64 1.000 5.030 RM000177 Propylene glycol 7.000 35.03 RM000181 Propylene glycol 2.675 13.46 RM000181 Purified water 35.00 175.2 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated:


010203040506070

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )







207


BATCH SUMMARY


Material % (w/w) Amount added (g) Rhodes # CP 0.05000 0.2498 RM000150 Propylene glycol 44.50 222.55 RM000181 Sodium citrate 0.05000 0.2490 RM000183 Citric acid 0.05000 0.2523 RM000185 Geleol® pastilles 5.000 25.01 RM000182 Cetostearyl alcohol 4.000 20.09 RM000184 White beeswax BP 0.6000 3.000 RM000142 Chlorocresol 0.07500 0.3765 RM000142 Gelot® 64 1.000 5.060 RM000177 Propylene glycol 7.000 35.10 RM000181 Propylene glycol 2.675 13.47 RM000181 Purified water 35.00 175.40 N/A Production equipment used: Water bath: Model NB-34980 Colora Ultra-Thermostat water bath Hotplate: Model RCH IKA-Combimag hotplate magnetic stirrer Homogenizer: Model 6-105 AF Virtis homogenizer Parameters evaluated :


0

10

20

30

40

50

60

0 2 4 6 8 10Time (h)0.5

Q (µ

g/cm

2 )


Mean SD CP content (%) 113.1 2.605 Viscosity (KcP) 11.43 0.05774 pH 5.15 0.0264





208

APPENDIX SIX

BATCH PRODUCTION RECORDS

Note that only one (1) batch production record for the CP cream formulations viz., production

record for Batch CP001 is included here. The batch production records for the other five (5)

batches, viz., Batches CP002, CP003, CP004, CP005 and CP006 are available on request.

209


BATCH PRODUCTION RECORD

Product name: Clobetasol 17-propionate cream Page: 1 of 6

Batch ID: CP001 Batch size: 200 g

MANUFACTURING APPROVALS

Batch record issued by__________________________ Date______________________

Master record issued by__________________________ Date______________________

210





MASTER FORMULA AND BATCH FORMULA

Item Material Quantity (% w/w)

Rhodes # Amount/ Batch

(g)

Amount dispensed

(g)

Dispensed by

Checkedby

1 CP 0.05000 RM000150 0.1000 0.1023 2 Propylene glycol 44.50 RM000181 89.00 89.04 3 Sodium citrate 0.05000 RM000183 0.100 0.1029 4 Citric acid 0.05000 RM000185 0.100 0.1026 5 Geleol® pastilles 5.000 RM000182 10.00 10.05 6 Cetostearyl alcohol 4.000 RM000184 8.000 8.053 7 White beeswax BP 0.6000 RM000142 1.200 1.020 8 Chlorocresol 0.07500 RM000186 0.1500 0.1508 9 Estol® 1474 1.000 RM000160 2.000 2.005 10 Propylene glycol 7.000 RM000181 14.00 14.10 11 Propylene glycol 2.675 RM000181 5.350 5.350 12 Purified water 35.00 N/A 70.00 70.07

211





EQUIPMENT VERIFICATION

Description Type Verified by Confirmed by Weighing scale Model AE-163 Mettler Hot plate Model RCH IKA-Combimag Water bath Model NB-34980 Colora Ultra-Thermostat Sonicator Model 8845-30 ultrasonic bath Homogenizer Model 6-105 AF Virtis Homoginizer Model-HO valve type

212





MANUFACTURING PROCEDURE

Step Procedure Time Done by Checked by 1 Weigh all the materials 2 Heat water (item 12) to 90˚C in a beaker. Bring

down the temperature to 60˚C. Dissolve sodium citrate (item 3) and citric acid (item 4) to a clear solution. Mix the resultant solution with propylene glycol (item 2). Maintain the temperature of the resultant aqueous phase (step 4) at 60˚C.

3 Melt Geleol® pastilles (item 5) and Estol® 1474 (item 9) together with cetostearyl alcohol (item 6), white beeswax BP (item 7) and chlorocresol (item 8) while stirring in a beaker previously heated to 75˚C using a hotplate. Cool the resultant oil phase to 60°C in a water bath and maintain the temperature at 60°C.

4 Mix CP (item 1) in propylene glycol (item 10) and sonicate for approximately twenty-five (25) minutes or until a clear solution is obtained. Heat the resultant drug phase to 50˚C and maintained at 50˚C.

5 Transfer the aqueous phase (step 2) to the oil phase (step 3) at 60˚C. Stir the mixture manually with a glass stirring rod for ten (10) minutes at 60˚C. Homogenize the mixture at 15,000 rpm for five (5) minutes at 60˚C.

6 Place the beaker containing the mixture (step 5) in a water bath at 20˚C and cool down the temperature of the resultant dispersed phase to 50˚C while stirring manually

213





MANUFACTURING PROCEDURE

Step Procedure Time Done by Checked by 7 Add the drug phase (step 4) to the dispersed phase

(step 7). Rinse the beaker previously containing the drug phase with propylene glycol (item 11) and add to the dispersed phase. Mix manually for ten (10) minutes at 50˚C. Homogenize the resultant cream at 15,000 rpm for five (5) minutes at 50˚C.

8 Cool the resultant cream (step 7) to 30˚C with continual manual stirring and with the manufacturing beaker placed in a water bath at 20˚C.

9 Pass the cooled formulation (step 8) through a valve type of homogenizer, to generate a smooth cream of improved consistency.

10 Pack the resultant cream (step 9) into 100 g ointment jars and store the cream at room temperature (22˚C) until required for further analysis.

214





SIGNATURE AND INITIAL REFERENCE Full Name (Print) Signature Initials Date

215

APPENDIX SEVEN

STABILITY TEST SUMMARY SHEET

216


STABILITY TEST SUMMARY SHEET

ACCELERATED STABILITY STUDIES

Product name: Clobetasol 17-propionate cream Manufacturer: Kasongo Wa Kasongo Address: Rhodes University, Faculty of Pharmacy, Division of Pharmaceutics, Grahamstown, South Africa Active ingredient: Clobetasol 17-propionate Dosage form: Cream

Batch ID Date of manufacture CP004 10-10-2006

Batch size Type of batch

500 g Experimental Samples tested (per batch): Four (4) samples Test period: Four (4) weeks

Storage/test conditions Temperature (˚C) Humidity (% RH)

40.0 25.0

RESULTS Chemical findings: • The percentage change from the initial amount of CP in the cream formulation after storage at 40 ±

2˚C and 25 ± 5% RH for 1, 2, 3, and 4 weeks was not statistically significant or relevant. • The percentage change from the initial apparent pH of the cream formulation after storage at 40 ± 2˚C

and 25 ± 5% RH for 1, 2, 3, and 4 weeks is not statistically significant or relevant Physical findings • No noticeable change in the organoleptic properties of the cream formulation in terms of appearance,

colour and odour over the entire stability test period. No signs of physical instability such as phase separation or cracking and bleeding of the cream base from the container at any of the sampling times.

• The percentage change from the initial apparent intrinsic viscosity of the CP cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for 1, 2, 3, and 4 weeks is not statistically significant or relevant.

• The percentage change from the initial CP in vitro release rate of CP release from the cream formulation after storage at 40 ± 2˚C and 25 ± 5% RH for weeks 1, 2, 3, and 4 was not statistically significant or relevant

Conclusions: The CP cream formulation (Batch CP004) was stable for the 1 month test period under investigation. Responsible Officer: Date:

217

REFERENCES

1. M.L. Gordon. The role of clobetasol propionate emollient 0.05% in the treatment of

patients with dry, scaly, corticosteroid-responsive dermatoses. Clinical Therapeutics 1998; 20(1):26-38.

2. J.D. Rosso, S.F. Friendlander. Corticosteroids: Options in the era of steroi-sparing therapy. Journal of American Academy of Dermatology 2005; 53(1):S50-S58.

3. J.Y. Fang, K.L. Shen, P.C.W. Huang, Y.H. Tsai. Evaluation of topical application of clobetasol 17-propionate from various cream bases. Drug Development and Industrial Pharmacy 1999; 25(1):7-14.

4. X. Huang, H. Tanojo, J. Lenn, H. Deng, L. Krochmal. A novel foam vehicle for delivery of topical corticosteroids. Journal of American Academy of Dermatology 2005; 53(1):S27-S38.

5. C. Purdon, J.M. Haigh, C. Surber, E.W. Smith. Foam drug delivery in dermatology. American Journal of Drug Delivery 2003; 1(1):71-75.

6. A.W. McKenzie, R.B. Stoughton. Method for comparing percutaneous absorption of steroids. Archives of Dermatology 1962; 86:608-610.

7. J.M. Haigh, I. Kanfer. Assessment of topical corticosteroid preparations: the human skin blanching assay. International Journal of Pharmaceutics 1984; 19(3):245-262.

8. E. Meyer, A.D. Magnus, J.M. Haigh, I. Kanfer. Comparison of the blanching activities of Dermovate, Betnovate and Eumovate creams and ointments. International Journal of Pharmaceutics 1988; 41:63-66.

9. B. Brazzini, N. Pimpinelli. New and established topical corticosteroids in dermatology; clinical pharmacology and therapeutic use. American Journal of Clinical Dermatology 2002; 3(1):47-58.

10. S. Dyderski, E. Grzeskowiak, A. Mrzyglod. Pharmaceutical availability of clobetasol 17-propionate from cream and ointment. Acta Poloniae Pharmaceutica-Drug Research 2001; 58(6):435-438.

11. M. Haramura, A. Tanaka, T. Kimoto, N. Hirayama. Crystal structure of clobetasol propionate. Analytical Sciences 2003; 19:x37-x38.

12. J.Y. Fang, C.L. Fang, K.C. Sung, H.Y. Chen. Effect of low frequency ultrasound on the in vitro percutaneous absorption of clobetasol 17-propionate. International Journal of Pharmaceutics 1999; 191(33):42.

13. J.C. Tsai, C.L. Cheng, Y.F. Tsai, H.M. Sheu, C.H. Chou. Evaluation of in vivo bioequivalence methodology for topical clobetasol 17-propionate based on

218

pharmacodynamic modeling using Chinese skin. Journal of Pharmaceutical Sciences 2004; 93(1):207-217.

14. Drug information fulltext [database online]. McEvoy GK, ed. Clobetasol propionate. American Society of Health-System Pharmacists, Bethesda, 2006. Updated Dec 2003. http://library.dialog.com/bluesheets/pdf/bl0229.pdf. Retrieved 24/06/2005.

15. Temovate (clobetasol propionate) cream and ointment 0.05% prescribing information, GlaxoSmithKline Consumer Healthcare LP. Updated 2002. http://us.gsk.com/products/assets/us_temovate.pdf. Retrieved 06/10/2006.

16. Temovate (clobetasol propionate) gel 0.05% prescribing information. GlaxoSmithKline Consumer HealthcareLP. Updated2002 http://us.gsk.com/products/assets/us_temovate_gel.pdf. Retrieved 8/02/2006.

17. Olux (Clobetasol propionate) foam 0.05% patient information. CCL Pharmaceuticals, Updated 2000. http://www.fda.gov/cder/foi/label/2000/21142lbl.pdf. Retrieved 02/08/2006.

18. S.C. Sweetman (editor): Martindale, "The Complete Drug Reference". 33rd Edition, The Pharmaceutical Press; London, 2002, pp. 1039-1065.

19. J.E.F. Reynolds (editor): Martindale, "The Extra Pharmacopoeia".30th Edition, The Pharmaceutical Press, London, 1993, pp. 712-740, 792, 851, 1028, 1108, 1355, 1406.

20. C. Dollery: Clobetasol propionate. In: C. Dollery (editor): Therapeutic Drugs. Churchill Livingstone; Edinburgh, 1991, pp. C272-C274.

21. A.C. Moffat, M.D. Osselton, B.Widdop (editors): Clarkes's Analysis of Drugs and Poisons: in pharmaceuticals, body fluids and post-mortem material. 3rd Edition, Pharmaceutical Press; London, 2003, pp. 821.

22. The British Pharmacopoeia incorporating the requirements of the 5th of the European Pharmacopoeia 2004 as amended by Supplements 5.1 and 5.2. The Stationery Office; London, 2005, pp. 506-507, 2749-2757, A167-A168, A272-A277.

23. The United States Pharmacopoeia incorporating "The National Formulary".The United States Pharmacopoeial Convention Inc.; Rockville, 2006, pp. 546-548, 2644-2646, 2682-2693, 3050-3053.

24. Data from SRC physprop database, CAS Number: 25122-46-7, Syracuse Research Corporation, 2004, http://www.syrres.com/esc/physdemo.htm. Retrieved 13/02/206.

25. S.F. Donovan, M.C. Pescatore. Method for measuring the logarithm of the octanol-water partition coefficient by using short octadecyl-poly(vinyl alcohol) high-performance liquid chromatography columns. Journal of Chromatography A 2002; 952(1-2):47-61.

219

26. J.J. Ellington, T.L. Floyd. Measuring octanol/water partition coefficients by the "slow-stirring" method. Environmental Research Brief, EPA/600/S-96/005, United States Environmental Protection Agency, Research and Development, Athens, 1995, pp. 1-4.

27. S.K. Poole, C.F. Poole. Separation methods for estimating octanol-water partition coefficients. Journal of Chromatography B 2003; 797(1-2):3-19.

28. J.T. Andersson, W. Schrader. A method for measuring 1-octanol-water partition coefficients. Analytical Chemistry 1999; 71(16):3610-3614.

29. M.S. Roberts, W.J. Pugh, J. Hadgraft. Epidermal permeability: Penetrant structure relationships. 2. The effect of h-bonding groups in penetrants on their diffusion through the stratum corneum. International Journal of Pharmaceutics 1996; 132(1-2):23-32.

30. R.M. Silverstein, G.C. Bassler, T.C .Morrill. Infrared spectrometry. In: R.M. Silverstein, G.C. Bassler, T.C. Morrill (editors): Spectrometric Identification of Organic Compounds. John Wiley & Sons, Inc.; New York, 1981, pp. 95-180.

31. E. Meyer. A critical evaluation of the human skin blanching assay and comparative bioavailability studies on topical corticosteroids preparations. Rhodes University, Grahamstown, 1998, pp. 1-282.

32. B.B. Michniak. Advances in topical corticosteroid therapy. International Journal of Cosmetic Sciences 1983; 5:277-291.

33. E.S. Snell. Mechanisms of glucocorticoid activity-a review. In: L. Wilson, R. Marks (editors): Mechanisms of topical corticosteroid activity-a Glaxo symposium.: Churchill Livingstone, Edinburgh, 1976, pp. 136-145.

34. D.A. Norris. Mechanisms of action of topical therapies and the rationale for combination therapy. Journal of the American Academy of Dermatology 2005; 53(1, Supplement 1):S17-S25.

35. E.S. Snell. The pharmacological properties of corticosteroids in relation to clinical efficacy. In: D.D. Munro, L. Wilson (editors): Steroid and the skin. Blackwell Scientific Publications and Glaxo Laboratories Ltd; London, 1976, pp. 15-23.

36. A. Prakash, P. Benfield. Topical mometasone, a review of its pharmacological properties and therapeutic use in the treatment of dermatlogical disorders. Drugs 1998; 55(1):145-163.

37. R.I. Scheinman, P.C. Cogswell, A.K. Lofquist, A.S.jr. Baldwin. Role of transcriptional activation of IκBα in mediation of immunosuppression by glucocorticoids. Science 1995; 270(5234):283-286.

38. N. Auphan, J.A. DiDonato, C. Rosette, A. Helmberg, M. Karin. Immunosuppression by glucocorticoids: Inhibition of NF-κB activity through induction of IκB synthesis. Science 1995; 270(5234):286-290.

220

39. S.R. Feldman. Relative efficacy and interchangeability of various clobetasol propionate vehicles in the management of steroid-responsive dermatoses. Current Therapeutic Research 2005; 66(3):154-171.

40. F.M. Keane, S.E. Munn, N.F. Taylor, A.W.P.D.U. Vivier. Unregulated use of clobetasol propionate. British Association of Dermatologists, British Journal of Dermatology 2001; 144:1095-1096.

41. A. Tosti, B.M. Piraccini, M. Pazzaglia, C. Vincenzi. Clobetasol propionate 0.05% under occlusion in the treatment of alopecia totalis/universalis. Journal of the American Academy of Dermatology 2003; 49(1):96-98.

42. E.W. Smith, H.I. Maibach. Percutaneous penetration enhancers; the fundamentals. In: E.W. Smith, H.I. Maibach (editors): Percutaneous Penetration Enhancers. CRC Press, Inc., New York 1995, pp. 1-4.

43. B.T. Gjertsen, J. Korsvold. Clobetasol propionate in psoriasis: monotherapy and in combinations. Journal of the European Academy of Dermatology and Venereology 1997; 9(Supplement 1):S173-S174.

44. S.C. Chattaraj, R.B. Walker. Penetration enhancer classification. In: E.W. Smith (editor): Percutaneous Penetration Enhancers. H.I. Maibach: CRC Press, Inc.; New York, 1995, pp. 5-20.

45. R.B. Walker, E.W. Smith. The role of percutaneous penetration enhancers. Advanced Drug Delivery Reviews 1996; 18(3):295-301.

46. S. Yateen, K.S. Dighe. High performance liquid chromatographic method for clobetasol propionate in dosage forms. Indian Drugs 1989; 27(2):132-133.

47. J.C. Reepmeyer, L.K. Revelle, I. Vidavsky. Detection of clobetasol propionate as an undeclared steroid in zinc pyritione formulations by high-performance liquid chromatography with rapid-scanning ultraviolet spectroscopy and mass spectrometry. Journal of Chromatography A 1998; 828:239-246.

48. L. Gagliardi, D. De Orsi, D. Tonelli. HPLC determination of clobetasol propionate in cosmetic products. Journal of Liquid Chromatography & Related Technologies 2000; 23(3):355-362.

49. Guidance for Industry: SUPAC-SS Non-sterile Semi-solid Dosage Forms. Scale-Up and Post-approval Changes: Chemistry, Manufacturing, and Controls; In Vitro Release Testing and In Vivo Bioequivalence Documentation. SUPAC-SS CMC 7, U.S. Department of Health and Human Services, Food and Drug Administration, Centre for Drug Evaluation and Research, Rockville, 1997, 1-36.

50. E.W. Smith, J.M. Haigh, I. Kanfer. A stability-indicating HPLC assay with on-line clean-up for betamathanose 17-valerate in topical dosage forms. International Journal of Pharmaceutics 1985; 27:185-192.

221

51. C.V. Garcia, A.R. Breier, B.M. Steppe, E.E.S. Schapoval, T.P. Oppe. Determination of dexamethasone acetate in cream by HPLC. Journal of Pharmaceutical and Biomedical Analysis 2003; 31(597):600.

52. R.J. Hamilton, P.A. Sewell. Introduction to high performance liquid chromatography. In: R.J. Hamilton, P.A. Sewell (editors): Introduction to high performance liquid chromatography. Chapman and Hall; New York, 1981, pp. 1-12.

53. R. Raghavan, J.C. Joseph. Chromatographic methods of analysis: high-performance liquid chromatography. In: J. Swarbrick, J.C. Boylan (editors): Encyclopedia of Pharmaceutical Technology. Marcel Dekker, Inc.; New York, 2002, pp. 414-425.

54. L.R. Snyder, J.J. Kirkland. Liquid-liquid chromatography. In: L.R. Snyder, J.J. Kirkland (editors): Introduction to modern liquid chromatography. John Wiley & Sons, Inc.; New York: 1979, pp. 323-348.

55. L.R. Snyder, J.J. Kirkland. Basic concepts and control of separation. In: L.R. Snyder, J.J. Kirkland (editors): Introduction to modern liquid chromatography. Jonh Wiley & Sons, Inc.; New York,1979, pp. 15-82.

56. L.R. Snyder, J.J. Kirkland. Bonded-phase chromatography. In: L.R. Snyder, J.J. Kirkland (editors): Introduction to modern liquid chromatography. John Wiley & Sons, Inc.; New York, 1979, pp. 269-322.

57. G.M. Pollack. Chromatographic Analysis. In: J. Swarbrick (editor): Encyclopedia of pharmaceutical technology. Marcel Dekker; New York,1987, pp. 419-445.

58. R.E. Majors. Practical operation of bonded-phase columns in high-performance liquid chromatography. In: C. Horvath (editor): High-performance liquid chromatography: Advances and perspectives. Academic Press; New York, 1980, pp. 75-111.

59. L.R. Snyder, J.J. Kirkland, J.L. Glajch. Getting started. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Practical HPLC method development. 2nd Edition. John Wiley & Sons, Inc; New York, 1997, p. 10.

60. M.M. Rahman, M. Takafuji, H. Ihara. Retention mechanism of L-glutamide-derived noncrystalline stationary phases in reversed-phase high-performance liquid chromatography and application for separation of nucleic acid constituents. Journal of Chromatography A 2006; 1119(1-2):105-114.

61. T.H Dzido, T.E. Kossowski, D. Matosiuk. Comparison of retention of aromatic hydrocarbons with polar groups in binary reversed-phase high-performance liquid chromatography systems. Journal of Chromatography A 2002; 947(2):167-183.

62. B. Mueller, Y. G.Anissimov, M.S. Roberts. Unexpected clobetasol propionate profile in human stratum corneum after topical application in vitro. Pharmaceutical Research 2003; 20:1835-1837.

222

63. L.R. Snyder, J.J. Kirkland, J.L. Glajch. Detection sensitivity and selectivity. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Practical HPLC method development. 2nd Edition. John Wiley and Sons, Inc.; New York: 1997, pp. 59-99.

64. L. Gagliardi, D. De Orsi, F. Manna, D. Tonelli. HPLC determination of clobetasol propionate in cosmetic products. Journal of Liquid Chromatography & Related Technologies 2000; 23(3):355-362.

65. L.R. Snyder, J.J. Kirkland, J.L. Glajch. The column. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Practical HPLC method development. 2nd Edition. John Wiley and Sons, Inc.; New York, 1997, pp. 174-230.

66. L.R.Snyder, J.J.Kirkland. The column. In: L.R.Snyder, J.J.Kirkland (editors): Introduction to modern liquid chromatography. New York: John Wiley & Sons, Inc.; 1979, pp. 168-245.

67. L.R. Snyder, J.J. Kirkland, J.L. Glajch. Non-ionic sample: reversed and normal-phase HPLC. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Practical HPLC method development. 2nd Edition. John Wiley and Sons, Inc.; New York, 1997, pp. 223-291.

68. N. Tanaka, K. Kimata, K. Hosoya, H. Miyanishi, T. Araki. Stationary phase effects in reversed-phase liquid chromatography. Journal of Chromatography A 1993; 656(1-2):265-287.

69. P. Zhuang, R.A. Thompson, T.P. O'Brien. A retention model for polar selectivity in reversed phase chromatography as a function of mobile phase organic modifier type. Journal of Liquid Chromatography & Related Technologies 2005; 28:1357-1365.

70. J.K. Fang, K.L. Shen, Y.B. Huang, P.C. Wu, Y.H. Tsai. Evaluation of Topical Application of Clobetasol 17-Propionate from Various Cream Bases. Drug Development and Industrial Pharmacy 1999; 25(1):7-14.

71. S. Kromidas. Optimization via column parameters: what works best? In: S. Kromidas (editor): More practical problem solving in HPLC. Wiley-VCH Verlag GmbH & Co. KGaA; Weinheim, 2005, pp. 14-16.

72. L.R. Snyder, J.J. Kirkland, J.L. Glajch. Basics of separation. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Practical HPLC method development. 2nd Edition. John Wiley and Sons, Inc.; New York, 1997, pp. 20-58.

73. C.M. Riley. Statistical parameters. In: C.M. Riley, T.W. Rosanske (editors): Development and validation of analytical methods. Elsevier Science Inc.; New York, 1996, pp. 44-45.

74. K. Hammarstrand. Internal standard in gas chromatography. Varian Instrument Applications 1976; 10:10-11.

223

75. J. Lindholm, M. Johansson, T. Fornstedt. Guidelines for analytical method development and validation of biotechnological synthesis of drugs: Production of a hydroxyprogesterone as model. Journal of Chromatography B 2003; 791(1-2):323-336.

76. L.R. Snyder, J.J. Kirkland, J.L. Glajch. Solvents. In: L.R. Snyder, J.J. Kirkland, J.L. Glajch (editors): Introduction to modern liquid chromatography. 2nd Edition John Wiley and Sons, Inc.; New York: 1979, pp. 246-268.

77. R.J. Hamilton, P.A. Sewell. Mobile phases in liquid chromatography. In: R.J. Hamilton, P.A. Sewell (editors): Introductio to high performance liquid chromatography. Chapman and Hall; New York, 1982, pp. 127-144.

78. Y. Sun, A.M. Stalcup. Mobile phase effects on retention on a new butylimidazolium-based high-performance liquid chromatographic stationary phase. Journal of Chromatography A 2006; 1126(1-2):276-282.

79. S. Kromidas. Methanol versus acetonitrile. In: S. Kromidas (editor): More practical solving in HPLC. Wiley-VCH Verlag GmbH & Co. KGaA; Weinheim: 2005, pp. 180-182.

80. L.C. Tan, P.W. Carr. Study of retention in reversed-phase liquid chromatography using linear solvation energy relationships: II: the mobile phase. Journal of Chromatography A 1998; 799(1-2):1-19.

81. G.A. Shabir. Validation of high-performance liquid chromatography methods for pharmaceutical analysis: Understanding the differences and similarities between validation requirements of the US Food and Drug Administration, the US Pharmacopeia and the International Conference on Harmonization. Journal of Chromatography A 2003; 987(1-2):57-66.

82. Reviewer Guidance: Validation of chromatographic methods. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research and Center for Veterinary Medicine. Rockville, 1994, pp. 1-30.

83. H. Rosing, W.Y. Man., E. Doyle, A. Bult, J.H. Beijnen. Bioanalytical liquid chromatographic method validation: a review of current practices and procedures. Journal of Liquid Chromatography and Related Technologies 2000; 23(3):329-354.

84. Guidance for industry: Bioanalytical methods validation. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research and Center for Veterinary Medicine.Rockville, 2001, pp. 1-22.

85. ICH Harmonised Tripartite Guidelines: validation of analytical methods: definitions and terminology. ICH Topic Q 2 A. International Conference on Harmonization, London, 1995, pp. 1-5.

224

86. ICH Harmonised Tripartite Guidelines: validation of analytical procedures: Methodology. ICH Topic Q 2 B. International Conference on Harmonization, London, 1996, pp. 1-9. 1996.

87. E. McGonogle. Assay validation and inter-laboratory transfer. In: C.M. Riley, T.W. Rosanske (editors): Development and validation of analytical methods. Elsevier Science Inc.; New York, 1996, pp. 3-13.

88. G.C. Hokanson. A life cycle approach to the development of analytical methods during pharmaceutical product development Part I: The initial method validation process. Pharmaceutical Technology 1994; 18(9):118-130.

89. G.S. Clarke. The validation of analytical methods for drug substances and drug products in UK pharmaceutical laboratories. Journal of Pharmaceutical and Biomedical Analysis 1994; 12(5):643-652.

90. G.P. Carr, J.C. Wahlich. A practical approach to method validation in pharmaceutical analysis. Journal of Pharmaceutical and Biomedical Analysis 1990; 8(8-12):613-618.

91. T.C. Paino, A.D. Moore. Determination of the LOD and LOQ of an HPLC method using four different techniques. Pharmaceutical Technology 1999; 23(10):82-90.

92. M.E. Swartz, I.S. Krull. Validation of chromatographic methods. Pharmaceutical Technology 1998; 22(3):104-118.

93. D. Dadgar, P.E. Burnett. Issues in evaluation of bioanalytical method selectivity and drug stability. Journal of Pharmaceutical and Biomedical Analysis 1995; 14(1-2):23-31.

94. A.R. Buick, M.V. Doig, S.C. Jeal, G.S. Land, R.D. McDowall. Method validation in the bioanalytical laboratory. Journal of Pharmaceutical and Biomedical Analysis 1990; 8(8-12):629-637.

95. D. Dadgar, P.E. Burnett, M.G. Choc, K. Gallicano, J.W. Hooper. Application issues in bioanalytical method validation, sample analysis and data reporting. Journal of Pharmaceutical and Biomedical Analysis 1995; 13(2):89-97.

96. S.R.J. Braggio, R.J. Barnaby, P. Grossi, M. Cugola. A strategy for validation of bioanalytical methods. Journal of Pharmaceutical and Biomedical Analysis 1996; 14(4):375-388.

97. U. Timm, M. Wall, D. Dell. A new approach for dealing with the stability of drugs in biological fluids. Journal of Pharmaceutical Sciences 1985; 74(9):972-977.

98. Guidance for industry: Analytical procedures and methods validations; Chemistry, manufacturing and controls documentation. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research. Rockville, 2000, pp. 1-33.

225

99. I. Taverniers, M. De Loose, E. Van Bockstaele. Trends in quality in the analytical laboratory. II. Analytical method validation and quality assurance. Trends in Analytical Chemistry 2004; 23(8):535-552.

100. P.M. Sathe, A.S. Raw, L.A. Ouderkirk, L.X. Yu, A.S. Husain. Drug product performance, in vitro. In: L. Shargel, I. Kanfer (editors): Generic drug product development: Solid oral dosage forms. Marcel Dekker, Inc.; New York, 2005, pp. 187-209.

101. W.A. Hanson. Overall considerations. In: W.A. Hanson (editor): Handbook of dissolution testing. Pharmaceutical Technology Publications; Oregon, 1982, pp. 1-11.

102. V.P Shah, J. Elkins, J. Hanus, C. Noorizadeh, J.P. Skelly. In vitro release of hydrocortisone from topical preparations and automated procedure. International Journal of Pharmaceutics 1989; 53(1):53-59.

103. S.C. Chattaraj, I. Kanfer. 'The “insertion cell”: a novel approach to monitor drug release from semi-solid dosage forms. International Journal of Pharmaceutics 1996; 133(1):59-63.

104. J. Klimundova, H. Sklenarova, U.F. Schaefer, P. Solich. Automated system for release studies of salicylic acid based on a SIA method. Journal of Pharmaceutical and BIomedical Analysis 2005; 37(5):893-898.

105. J. Klimundova, D. Satinsky, H. Sklenarova, P. Solich. Automation of simultaneous release tests of two substances by sequential injection chromatography coupled with a Franz cell. Talanta 2006; 69(3):730-735.

106. P.R. Rege, V.D. Vilivalam, C.C. Collins. Development in release testing of topical dosage forms: use of the Enhancer CellTM with automated sampling. Journal of Pharmaceutical and BIomedical Analysis 1998; 17(8):1225-1233.

107. W. Liebenberg, E. Engelbrecht, A. Wessels, B. Evarakonda, W. Yang, M.M. De Villie. A comparative study of the release of active ingredients from semisolid cosmeceuticals measured with Franz, Enhancer or flow-through cell diffusion apparatus. Journal of Food and Drug Analysis 2004; 12(1):19-28.

108. K.D. Thakker, W.H. Chern. Development and validation of in vitro release tests for semisolid dosage forms-case study. Dissolution Technologies 2003; 10(5):10-16.

109. V.P. Shah, C.R. Behl, G. Flynn, W.I. Higuchi, H. Schaefer. Principles and criteria in the development and optimization of topical therapeutic products. International Journal of Pharmaceutics 1992; 82(1):21-28.

110. S.C. Chattaraj, J. Swarbrick, I. Kanfer. A simple diffusion cell to monitor drug release from semi-solid dosage forms. International Journal of Pharmaceutics 1995; 120(1):119-124.

226

111. M. Siewett, J. Dressman, C.K. Brown, V.P. Shah. FIP/AAPS Guidelines to dissolution/in vitro release testing of novel/special dosage forms. American Association of Pharmaceutical Scientists PharmSciTech 2003; 4(1):1-10.

112. G.L. Flynn, V.P. Shah, S.N. Tenjarla, M. Corbo, D. DeMagistris, T.G. Feldman, et al,. Assessment of value and applications of in vitro testing of topical dermatological drug products. Pharmaceutical Research 1999; 16(9):1325-1329.

113. J.L. Zatz, J.D. Segers. Techniques for measuring in vitro release from semisolids. Dissolution Technologies 1998; 5(2):3-13.

114. R. Pillai, V.P. Shah, C.F. Albrecht, S.C. Chattaraj, G. Flynn. Release of hydrocortisone from a cream matrix: dependency of release on suspension concentration and measurement of solubility and diffusity. Pharmaceutical Development and Technology 2001; 6(3):373-384.

115. M. Corbo, T.W. Shultz, G.K. Wong, G.A. Van Buskirk. Development and validation of in vitro release testing methods for semisolid formulations. Pharmaceutical Technology 1993; 17(9):112-128.

116. P.P. Sanghvi, C.C. Collins. Comparison of diffusion studies of hydrocortisone between the Franz cell and the Enhancer cell. Drug Development and Industrial Pharmacy 1993; 19(13):1573-1585.

117. V.P Shah, J. Elkins, S.Y. Lam, J.P. Skelly. Determination of in vitro drug release from hydrocortisone creams. International Journal of Pharmaceutics 1989; 53(1):53-59.

118. M.D. Vlachou, D.M. Rekkas, P.P. Dallas, N.H.Choulis. Development and in vitro evaluation of griseofulvin gels using Franz diffusion cells. International Journal of Pharmaceutics 1992; 82(1-2):47-52.

119. L.J. Zatz. Drug release from semisolids: Effects of membrane permeability on sensitivity to product parameters. Pharmaceutical Research 1995; 12(5):787-789.

120. T.X. Viegas, L.L.V. Winkle, P.A. Lehman, S.F. Franz, T.J. Franz. Evaluation of creams and ointments as suitable formulations for peldesine. International Journal of Pharmaceutics 2001; 219(1-2):73-80.

121. V.P. Shah, J. Elkins, R.L. Williams. Evaluation of the test system used for in vitro release of drugs for topical dermatological drug products. Pharmaceutical Development and Technology 1999; 4(3):377-385.

122. J.P. Skelly, V.P. Shah, H.I. Maibach, R.H. Guy, R.C. Wester, G. Flynn, et al., FDA and AAPS report of the workshop on principles and practices of in vitro percutaneous penetration studies: relevance to bioavailability and bioequivalence. Pharmaceutical Research 1987; 4(3):265-267.

227

123. I. Csoka, E. Csanyi, G. Zapantis, E. Nagy, A. Feher-Kiss, G. Horvath, et al., In vitro and in vivo percutaneous absorption of topical dosage forms: case studies. International Journal of Pharmaceutics 2005; 291(1-2):11-19.

124. E.W. Smith, J.M. Haigh. In vitro diffusion cell design and validation II. Temperature, agitation and membrane effects on betamethasone 17-valerate permeation. Acta Pharmareutica Nordica 1992; 4(3):171-178.

125. V.P. Shah, J. Elkins, S. Shaw, R. Hanson. In vitro release: comparative evaluation of vertical diffusion cell system and automated procedure. Pharmaceutical Development and Technology 2003; 8(1):97-102.

126. M.H. Fares, L.J Zatz. Measurement of drug release from topical gels using two types of apparatus. Pharmaceutical Technology 1995; 19(1):53-58.

127. P.R. Keshary, Y.W. Chien. Mechanisms of transdermal controlled nitroglycerin administration (I): development of a finite-dosing skin permeation system. Drug Development and Industrial Pharmacy 1984; 10(6):883-913.

128. P. Gupta, S. Garg. Recent advances in semisolid dosage forms for dermatological application. Pharmaceutical Technology 2002; 26(3):144-162.

129. S.C. Pefile, E.W. Smith, C.F. Albrecht, P.B. Kruger. Release of rooperol tetra-acetate from topical bases: in vitro studies using silicone membrane. International Journal of Pharmaceutics 1998; 161(2):237-243.

130. E.J. Ricci, L.O. Lunardi, D.M.A. Nanclares, J.M. Marchetti. Sustained release of lidocaine from Poloxamer 407 gels. International Journal of Pharmaceutics 2005; 288(2):235-244.

131. B.W. Barry. Methods for Studying Percutaneous Absorption. In: B.W. Barry (editor): Dermatological Formulations-Percutaneous Absorption. Marcel Dekker, Inc.; New York: 1983, pp. 234-295.

132. T.J. Franz. On the relevance of in vitro data. Journal of Investigative Dermatology 1975; 64:190-195.

133. T.J. Franz. The finite dose technique as a valid in vivo model for the study of percutaneous absorption. Current Problems in Dermatology 1978; 7:58-68.

134. M. Dias, A. Farinha, E. Faustino, J. Hadgraft, J. Pais, C. Toscano. Topical delivery of caffeine from some commercial formulations. International Journal of Pharmaceutics 1999; 182(1):41-47.

135. L. Montenegro, A.M. Panico, A. Ventimiglia, F.P. Bonina. In vitro retinoic acid release and skin permeation from different liposome formulations. International Journal of Pharmaceutics 1996; 133(1):89-96.

228

136. L.G. Flynn, P.R. Keshary, Y.C. Huang. Comparative controlled skin permeation of nitroglycerin from markted transdermal delivery systems. Journal of Pharmaceutical Sciences 1983; 72(8):968-970.

137. T. Higuchi. Mechanism of sustained-action medication. Theoretical analysis of rate of release of solid drugs dispersed in solid matrices. Journal of Pharmaceutical Sciences 1963; 52(12):1145-1149.

138. W.I. Higuchi. Analysis of data on the medicament release from ointments. Journal of Pharmaceutical Sciences 1962; 51(8):802-804.

139. S.C. Owen. Alcohol. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, DC, 2003, pp. 13-15.

140. P.J. Weller. Propylene glycol. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association, Washington, DC, 2003, pp. 521-523.

141. A. Rolland, G. Demichelis, J. Jamoulle, B. Shroot. Influence of formulation, receptor fluid and occlusion on in vitro release from topical dosage forms using an automated flow-through diffusion cell. Pharmaceutical Research 1992; 9(1):82-86.

142. V.P. Shah, J.S. Elkins. In vitro release from corticosteroid ointments. Journal of Pharmaceutical Sciences 1995; 84(9):1139-1140.

143. J.M. Haigh, E.W. Smith. The selection and use of natural and synthetic membranes for in vitro diffusion experiments. European Journal of Pharmaceutical Sciences 1994; 2(5-6):311-330.

144. T. Higuchi. Rate of release of medicaments from ointment bases containing drugs in suspension. Journal of Pharmaceutical Sciences 1961; 50(10):874-875.

145. K. Welin-Berger, J.A.M. Neelissen, B. Bergenstahl. The effect of rheological behaviour of a topical anaesthetic formulation on the release and permeation rates of the active compound. European Journal of Pharmaceutical Sciences 2001; 13(3):309-318.

146. B.W. Barry. Basic principles of diffusion through membranes. In: B.W. Barry (editor): Dermatological Formulations, Percutaneous Absorption. Mercel Dekker, Inc.; New York, 1983, pp. 49-94.

147. A. Karim, G. Pillai, T.T. Ziqubu-page, M.H. Cassimjee, M.S. Morar. Potential savings from generic prescribing and generic substitution in South Africa. Health Policy and Planning 1996; 11(2):198-205.

148. D. Boyce, G. Bartlett. The maximum medical aid price programme: a review of the concept and its ability to reduce expenditure on medicines. South African Medical Journal 1990; 78(4):147-151.

229

149. D.R. Scott, W.D. Reekie. Savings from generic drug substitution in the RSA- is its cost justified? South African Medical Journal 1987; 71(3):314-316.

150. G. Hukins, D. Boyce. Containing medicines costs in the private sector-the example of a medical benefit scheme. South African Medical Journal 1990; 78(5):152-157.

151. L. Shargel, I. Kanfer. Introduction to generic drug product development. In: L. Shargel, I. Kanfer (editors): Generic Drug Product Development: Solid Oral Dosage Forms. Marcel Dekker; New York, 2005, pp. 1-16.

152. 9 countries from the WHO African Region meet to analyse data from their medicines prices surveys 2006, pp. 1-2. http://www.who.int/medicines/areas/access/AFROmedicinesPricesShortArticle.pdf. Retrieved 01/11/2006

153. M.N.G. Dukes. Accountability of the pharmaceutical industry. The Lancet 2002; 360(11):1682-1684.

154. J.K. Wang'ombe, G.M. Mwabu. Economics of essential drugs schemes: the perspectives of the developing countries. Social Science and Medicine 1987; 25(6):625-630.

155. G.G. Djolov. Savings from generic drug substitution in South Africa: an arithmetical view. South African Medical Journal 2003; 93(8):583-584.

156. Medicines and Related Susbstances Act 101 of 1965 (Act 101 of 1965) as amended. PSSA Pharmacy Law Compendium. LexisNexix Butterworths; Johannesburg, 2003, pp. MAE-30.

157. Office of generic drugs. The United States Food and Drug and Administration's Office of Generic Drug. 2006 http://www.fda.gov/cder/ogd/index.htm. Retrieved 01/11/2006.

158. The European Agency for the Evaluation of Medicinal Products (EMEA). http://www.emea.eu.int/. Retrieved 08/11/2006.

159. The Medical Control Council. http://www.mccza.com/main.asp. Accessed 08/11/2006.

160. D. Henry, J. Lexchin. The pharmaceutical industry as a medicines provider. The Lancet 2002; 360(9345):1590-1595.

161. The pharmaceutical industry in figures. The European Federation of Pharmaceutical Industries and Associations (EFPIA). http://www.efpia.org/6_publ/infigures2005.pdf, 2005, pp 1-28. Retrieved 08/11/2006.

162. S.T. Hartshorne. Dermatological disorders in Johannesburg, South Africa. Clinical and Experimental Dermatology 2003; 28(6):661-665.

230

163. Dermatologic disorders. In: M.H. Beers, R. Berkow (editors): The Merck Manual of Diagnosis and Therapy. 17th Edition, Whitehouse Station, Merck Research Laboratories, 1999, pp. 777-846.

164. A. Billich, H. Aschauer, A. Aszodi, A. Stuetz. Percutaneous absorption of drugs used in atopic eczema: pimecrolimus permeates less through skin than corticosteroids and tacrolimus. International Journal of Pharmaceutics 2004; 269(1):29-35.

165. Pharmaceutical Price Regulation Scheme (PPRS): The study into the extent of competition in the supply of branded medicines to the NHS. The United Kingdom Department of Health [Component3]. London, 2006, pp. 121-143. http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/PublicationsPAmpGBrowsableDocument/fs/en?CONTENT_ID=4099371&MULTIPAGE_ID=5024085&chk=dwu6lz. Retrieved 08/11/2006.

166. Corticosteroids, dermatological preparations. In: C.J. Gibbon (editor): South African Medicines Formulary. Pinelands: South African Medical Association: Health and Medical Publishing Group; Cape Town, 2003, pp. 188-190.

167. L.V. Allen J, N.G. Popovich, H.C. Ansel. Semi-solid dosage forms and transdermal systems: ointments, creams, and gels. In: L.V. Allen Jr, N.G. Popovich, H.C. Ansel (editors): Ansel's Pharmaceutical Dosage Forms and Drug Delivery Systems. Lippincott Williams & Wilkins; Baltimore, 2005, pp. 276-297.

168. L. Buhse, R. Kolinski, B. Westenberger, et a.,l. Topical drug classification. International Journal of Pharmaceutics 2005; 295(1-2):101-112.

169. N. Realdon, F. Perin, M. Morpurgo, E. Ragazzi. Influence of processing conditions in the manufacture of o/w creams: I. Effect on dispersion grade and rheological characteristics. Il Farmaco 2002; 57(5):341-347.

170. U.T. Lashmar, J.P. Richardson, A. Erbod. Correlation of physical parameters of an oil in water emulsion with manufacturing procedures and stability. International Journal of Pharmaceutics 1995; 125(2):315-325.

171. K.A. Walters, K.R. Brain. Dermatological formulation and transdermal systems. In: K.A. Walters (editor): Dermatological and Transdermal Formulations.: Mercel Dekker, Inc.; New York , 2002, pp. 319-399.

172. B.W. Barry. Formulation of dermatological vehicles. In: B.W. Barry (editor): Dermatological Formulations: Percutaneous Absorption. Mercel Dekker, Inc.; New York, 1983, pp. 296-350.

173. G.M. Eccleston. Functions of mixed emulsifiers and emulsifying waxes in dermatological lotions and creams. Colloids and Surfaces A: Physicochemical and Engineering Aspects 1997; 123-124:169-182.

231

174. M. Korhonen, J. Lehtonen, L. Hellen, J. Hirvonen, J. Yliruusi. Rheological properties of three component creams containing sorbitan monoesters as surfactants. International Journal of Pharmaceutics 2002; 247(1-2):103-114.

175. H. Okochi, M. Nakano. Preparation and evaluation of w/o/w type emulsions containing vancomycin. Advanced Drug Delivery Reviews 2000; 45(1):5-26.

176. T. Hino, Y. Kawashima, S. Shimabayashi. Basic study for stabilization of w/o/w emulsion and its application to transcatheter arterial embolization therapy. Advanced Drug Delivery Reviews 2000; 45(1):27-45.

177. M. Nakano. Places of emulsions in drug delivery. Advanced Drug Delivery Reviews 2000; 45(1):1-4.

178. S. Van der Graaf, C.G.P.H. Schroen, R.M. Boom. Preparation of double emulsions by membrane emulsification: a review. Journal of Membrane Science 2005; 251(1-2):7-15.

179. S. Tamilvanan. Oil-in-water lipid emulsions: implications for parenteral and ocular delivering systems. Progress in Lipid Research 2004; 43(6):489-533.

180. F.V. DiStefano, O.M. Shaffer, M.S. El-Aasser, J.W. Vanderhoff. Multiple oil-in-water-in-oil emulsions of extremely fine droplet size. Journal of Colloid and Interface Science 1983; 92(1):269-272.

181. M. Korhonen. Rheological properties of pharmaceutical creams containing sorbitan fatty acid ester surfactant. University of Helsink Helsinki, 2004, 1-58. http://ethesis.helsinki.fi/julkaisut/mat/farma/vk/korhonen/. Retrieved 10/12/2006.

182. M. Billany. Suspensions and emulsions. In: M.E. Aulton (editor): Pharmaceutics: The Science of Dosage Form Design. Churchill Livingstone; New York: 2002, pp. 334-359.

183. G.L. Flynn. Topical Drug Absorption and Topical Pharmaceutical Systems. In: G.S. Banker, C.T. Rhodes (editors): Modern Pharmaceutics. Marcel Dekker, Inc.; New York, 1990, pp. 298-302.

184. T. Tadros. Application of rheology for assessment and prediction of the long-term physical stability of emulsions. Advances in Colloid and Interface Science 2004; 108-109:227-258.

185. H.W. Yarranton, J.H. Masliyah. Numerical Simulation of Ostwald Ripening in Emulsions. Journal of Colloid and Interface Science 1997; 196(2):157-169.

186. P. Taylor. Ostwald ripening in emulsions. Advances in Colloid and Interface Science 1998; 75(2):107-163.

232

187. H. Masmoudi, Y. Le Dreau, P. Piccerelle, J. Kister. The evaluation of cosmetic and pharmaceutical emulsions aging process using classical techniques and a new method: FTIR. International Journal of Pharmaceutics 2005; 289(1-2):117-131.

188. M.M. Robins, A.D. Watson, P.J. Wilde. Emulsions: creaming and rheology. Current Opinion in Colloid & Interface Science 2002; 7(5-6):419-425.

189. M.M. Robins. Emulsions: creaming phenomena. Current Opinion in Colloid & Interface Science 2000; 5(5-6):265-272.

190. P. Taylor. Ostwald ripening in emulsions. Colloids and Surfaces A: Physicochemical and Engineering Aspects 1995; 99(2-3):175-185.

191. G.A. Buskirk, V.P. Shah, D. Adair, et al., Workshop III report: scale-up of liquid and semi-solid disperse systems. Pharmaceutical Research 1994; 11(8):1216-1219.

192. U.T. Lashmar, J. Beesley. Correlation of rheological properties of an oil-in-water emulsion with manufacturing procedures and stability. International Journal of Pharmaceutics 1993; 91(1):59-67.

193. C. Surber, A.F. Davis. Bioavailability and bioequivalence of dermatological formulations. In: K.A. Walters (editor): Dermatological and Transdermal Formulations. Marcel Dekker, Inc.; New York, 2002, pp. 401-498.

194. G. Yosipovitch, H.I. Maibach. Skin surface pH: a protective acid mantle. Cosmetics and Toiletries 1996; 111(12):101-102.

195. I. Kanfer, R.B. Walker, P. Persicaner. Experimental formulation development. In: L. Shargel, I. Kanfer (editors): Generic Drug Product Development: Solid Oral Dosage Forms. Mercel Dekker; New York, 2005, pp. 53-93.

196. Patient information leaflet: Dermivate® cream: clobetasol propionate, GlaxoSmithKline plc.http://emc.medicines.org.uk/emc/assets/c/html/displaydoc.asp?documentid=3829. Retrieved 12/12/2006.

197. British Pharmaceutical Codex. The Pharmaceutical Press; London, 1973, pp. 40, 88, 100, 114, 413, 449. .

198. Propylene glycol. In: The British Pharmacopoeia. The Pharmaceutical Press, London, 2005, pp. 1680-1681.

199. G.E. Amidon. Sodium citrate dihydrate. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients.: American Pharmaceutical Association; Washington, D.C. , 2003, pp. 560-562.

200. G.E. Amidon. Citric acid monohydrate. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, D.C., 2003, pp. 158-160.

233

201. A.K. Taylor. Glyceryl monostearate. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, D.C., 2003, pp. 264-266.

202. G. Frunzi, B. Sarsfield. Cetostearyl alcohol. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, D.C., 2003, pp. 125-126.

203. A.H. Kibbe. Wax, white. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, DC, 2003, pp. 687-688.

204. P.J. Weller. Chlorocresol. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, DC 2003, pp. 144-146.

205. Uniqema: Estol® 1474. http://www.uniqema.com/irj/portal?NavigationTarget=navurl://8975db69ba619ede9ec8a256b362aded. Retrieved 05/11/2006.

206. Masso: personal care ingredients-emulsifiers. http://www.cqmasso.com/fcr/cosmetica/catalogo/Type_of_Product/General%20Catalogue%20-%20Emulsifiers%20(M&Q).pdf. Retrieved 05/11/2006.

. 207. SpecialChem: innovation and solutions in cosmetics: Ritapro® 200. http://www.specialchem4cosmetics.com/common/COS/services/ingredients/displayproduct.aspx?or=docindex&menu=true&id=5149. Retrieved 05/11/2006.

. 208. Gattefossé: dermal route excipients; enhancing pharmaceutical performance. 3rd Edition, Saint-Priest Cedex, 1999. Production information sheet.

209. Chemical drug information: Emulcire® 61 WL.ttp://www.kreglinger-europe.com/eng/product?21202. Kreglinger Europe.

210. S.K. Niazi. Formulations of semi-solid drugs: Clobetasol propionate cream. In: S.K. Niazi (editor): Handbook of Pharmaceutical Manufacturing Formulations: Semi-solid products. CRC Press; Boca Raton, 2004, pp. 136-137.

211. L.M. Spalton. Formulation and preparation of emulsions. In: L.M. Spalton (editor): Pharmaceutical Emulsions and emulsifying agents. The chemist and Druggist; London: 1959, pp. 29-49.

212. F. Niellound, G. Marti-Mestres, J.P. Laget, C. Fernandez, H. Maillols. Emulsion formulations: study of the influence of parameters with experimental designs. Drug development and Industrial Pharmacy 1996; 22(2):156-166.

234

213. B.W. Barry. Basic principles of diffusion through membranes. In: B.W. Barry (editor): Dermatological Formulations, Percutaneous Absorption. Mercel Dekker, Inc.; New York: 1983, pp. 49-94.

214. K.Welin-Berger J.A.M. Neelissen, B. Bergenstahl. The effect of rheological behaviour of a topical anaesthetic formulation on the release and permeation rates of the active compound. European Journal of Pharmaceutical Sciences 2001; 13(3):309-318.

215. B. Kommanaboyina, C.T. Rhodes. Trends in stability testing with emphasis on stability during distribution and storage. Drug Development and Industrial Pharmacy 1999; 25(7):857-868.

216. K. Komka, S.Kemeny. A modified error model for the assessment of stability of pharmaceutical products. Chemometrics and Intelligent Laboratory Systems 2004; 72(2):161-165.

217. WHO expert committee on specifications for pharmaceutical preparations (WHO technical report, series 863). 34, 65-76. 1996. Geneva, World Health Organization.

218. S. Singh. Drug stability testing and shelf-life determination according to international guidelines. Pharmaceutical Technology 1999; 23(6):68-88.

219. B.R.Matthews. Regulatory aspects of stability testing in Europe. Drug Development and Industrial Pharmacy 1999; 25(7):831-856.

220. P.K. Bhattacharyya. Drug stability. In: L. Shargel, I. Kanfer (editors): Generic Drug Product Development: Solid Oral Dosage Forms. Marcel Dekker; New York, 2005, pp. 137-172.

221. ICH harmonised tripartite guideline: stability testing of new drug substances and products: Q1A (R2). CPMP/ICH/2736/99. International Conference on Harmonisation, London, 2003, pp.1-20.

222. C.T. Rhodes. Introductory overview. In: J.T. Carstensen, C.T. Rhodes (editors): Drug Stability: Principles and Practices, Third Edition, Revised and Expanded. Marcel Dekker; New York, 2000, pp. 2-18.

223. J.J. Verner. Formulation and dissolution assessment of a novel repeat action tablet containing a decongestant and an antihistamine. Rhodes University. Grahamstown, 2003, pp. 136-137.

224. H. Forcinio. Choosing a blister material. Pharmaceutical Technology 2000; 24(5):26-30.

225. S. Singh. Stability test storage conditions for zones III and IV: some unresolved issues. Pharmaceutical Technology 1999; 23(10):130-142.

235

226. D. Pecanac, F. Van Houtte, E. Roets, J. Hoogmartens. A stability study of ticlopidine products from 18 countries. Drug Development and Industrial Pharmacy 2000; 26(4):391-401.

227. Guidance for industry: guideline for submitting documentation for the stability of human drugs and biologics. Food and Drug Administration, Rockville, 1987, 1-56.

228. J.D. Haynes. Worldwide virtual temperatures for pharmaceutical products stability testing. Journal of Pharmaceutical Sciences 1971; 60(6):927-919.

229. S.C. Owen. Sodium chloride. In: R.C. Rowe, P.J. Sheskey, P.J. Weller (editors): Handbook of Pharmaceutical Excipients. American Pharmaceutical Association; Washington, D.C., 2003, pp. 556-559.

230. J.T. Carstensen. Physical testing. In: J.T. Carstensen, C.T. Rhodes (editors): Drug Stability: Principles and Practices: Third Edition, Revised and Expanded. Mercel Dekker; New York, 2000, pp. 261-328.

231. N.D. Ramirez-Beltran, L.A. Olivares. Statistical analysis of drug stability. Computers & Industrial Engineering 1999; 37(1-2):351-354.

232. M.E. Gil-Alegre, J.A. Bernabeu, M.A. Camacho, A.I. Torres-Suarez. Statistical evaluation for stability studies under stress storage conditions. Il Farmaco 2001; 56(11):877-883.

233. B. Kommanaboyina, C.T. Rhodes. Effects of temperature excursions on mean kinetic temperature and shelf-life. Drug Development and Industrial Pharmacy 1999; 25(12):1301-1306.

DEVELOPMENT AND IN VITRO EVALUATION OF A CLOBETASOL … · 2017-02-12 · 1. To develop, optimize...

Documents

Transcript of DEVELOPMENT AND IN VITRO EVALUATION OF A CLOBETASOL … · 2017-02-12 · 1. To develop, optimize...